Journal of Management, Informatics and Human Resources Volume 57, Issue 4, November 2024 ISSN 1318-5454 Revija za management, informatiko in kadre Organizacija (Journal of Management, Informatics and Organizacija (Revija za management, informatiko in Human Resources) is an interdisciplinary peer-reviewed človeške vire) je interdisciplinarna recenzirana revija, ki journal which is open to contributions of high quality, from objavlja visoko kakovostne prispevke z vseh vidikov, ki so any perspective relevant to the organizational phenomena. pomembni za organizacijske procese in strukture. Revija je zasnovana tako, da spodbuja zanimanje za The journal is designed to encourage interest in all matters različne vidike v zvezi z organizacijskimi vedami in je relating to organizational sciences and is intended to ap- namenjena tako akademski kot strokovni skupnosti. peal to both the academic and professional community. In Revija objavlja izvirne članke, ki spodbujajo empirično, particular, journal publishes original articles that advance teoretično in metodološko razumevanje teorij in konceptov the empirical, theoretical, and methodological understand- managementa in organizacije. Pozdravljamo tudi ing of the theories and concepts of management and or- prispevke iz drugih znanstvenih disciplin, ki spodbujajo ganization. The journal welcomes contributions from other nove koncepte v organizacijski teoriji in praksi. Objavljamo scientific disciplines that encourage new conceptualiza- članke, ki analizirajo organiziranost z različnih vidikov, so tions in organizational theory and management practice. usmerjeni na organizacije različnih velikosti in iz različnih sektorjev, na enote, ki sestavljajo organizacije, in na mreže, We welcome different perspectives of analysis, including v katere so organizacije vpete. the organizations of various sizes and from various branch- es, units that constitute organizations, and the networks in Teme so pokrivajo predvsem naslednja področja: which organizations are embedded. • organizacijska teorija, upravljanje, razvoj in Topics are drawn, but not limited to the following areas: organizacijsko vedenje; • organizational theory, management, development, • management človeških virov (kot so organizacija in and organizational behaviour; razvoj zaposlenih, vodenje, ustvarjanje vrednosti s pomočjo človeških virov, organizacijski pojavi na • human resources management (such as organization delovnem mestu itd.); & employee development, leadership, value creation through HRM, workplace phenomena etc.); • vodstveni in podjetniški vidiki izobraževanja; • managerial and entrepreneurial aspects of education; • poslovni informacijski sistemi (kot so digitalno poslovanje, sistemi za podporo odločanju, poslovna • business information systems (such as digital business, analitika itd.); decision support systems, business analytics etc.); • podjetniški inženiring (npr. organizacijsko oblikovanje, • enterprise engineering (e.g., organizational upravljanje poslovnih procesov, paradigme design, business process management, enterprise preoblikovanja podjetij itd.); transformation paradigms etc.); • članki, ki analizirajo organizacijsko uspešnost in • papers that analyse and seek to improve organizational prizadevanja za izboljšanje le-te. performance. Organizacija, Volume 57, Issue 4 November 2024 Contents 4/2024 RESEARCH PAPERS 319 Pavel PELECH, An Examination of Generational Differenc- Jaroslava DĚDKOVÁ es in the Sharing Economy: Understand- ing the Motivations of Idle Asset Owners and Their Impact on Peer-to-Peer Platform Management 333 Eva JEREB, The Use of Artificial Intelligence among Marko URH Students in Higher Education 346 Aykut ARSLAN, Multilevel Investigation of Leadership Serdar YENER, Prototype Perception: Political Behavior in Fatma KORKMAZ, Relation to Effectiveness and Trust Murat Sakir EROGUL 363 Barbara DORIĆ, Government Effectiveness in the Dinko PRIMORAC, Petroleum Sector: Two-step Analysis Mirjana PEJIĆ BACH Combining Linear Regression and Artificial Neural Networks 379 Božidar LENARČIČ, Using the Nominal Group Technique to Armand FAGANEL Design an Appropriate Communication Strategy Among Strategic Procurement and Internal Stakeholders 392 Gita ŠAKYTĖ-STATNICKĖ The Effect of Brand Association and Brand Image on the Brand Awareness of Lithuanian Companies of Digital Products Editorial office: University of Maribor, Faculty of Organizational Science, Založba Moderna Organizacija, Kidriceva 55a, 4000 Kranj, Slovenia Tel.: +386-4-2374297, E-mail: organizacija.fov@um.si, URL: https://organizacija.fov.um.si. Organizacija is co-sponsored by the Slovenian Research Agency. Published quarterly. Full text of articles are available at https://sciendo.com/journal/ORGA. Papers for publication in Organizacija can be submitted via journal website at https://organizacija.fov.um.si/index.php/organizacija/information/authors. Before submission authors should consult. Guidelines available at https://sciendo.com/journal/ORGA, tab “for Authors”. You can contact the editorial via e-mail: organizacija.fov@um.si or maja.mesko@um.si Articles are currently abstracted/indexed in: Cabell’s Directory, CEJSH (The Central European Journal of Social Sciences and Humanities), Celdes, Clarivate Analytics - Emerging Sources Citation Index (ESCI), CNPIEC, Die Elektronische Zeitschriftenbibliothek, DOAJ, EBSCO - TOC Premier, EBSCO Discovery Service, ECONIS, Ergonomics Abstracts, ERIH PLUS, Google Scholar, Inspec, International Abstracts in Operations Research, J-Gate, Microsoft Academic Search, Naviga (Softweco), Primo Central (ExLibris), ProQuest - Advanced Pol mers Abstracts, ProQuest - Aluminium Industry Abstracts, ProQuest - Ceramic Abstracts/World Ceramics Abstracts, ProQuest - Composites Industry Abstracts, ProQuest - Computer and Information Systems Abstracts, ProQuest - Corrosion Abstracts, ProQuest - Electronics and Co mmunications Abstracts, ProQuest - Engineered Materials Abstracts, ProQuest - Mechanical & Transportation Engineering Abstracts, ProQuest - METADEX (Me tals Abstracts), ProQuest - Sociological Abstracts, ProQuest - Solid State and Superconductivity Abstracts, Research Papers in Economics (RePEc), SCOPUS, Summon (Serials Solutions/ProQuest), TDOne (TDNet), TEMA Technik und Management, WorldCat (OCLC). Organizacija, Volume 57 Issue 4, November 2024 EDITOR / UREDNIK EDITORIAL BOARD / Nataša Petrović UREDNIŠKI ODBOR REVIJE University of Belgrade, Serbia Maja Meško University of Maribor, Faculty of Hossein Arsham, Tetyana Pimonenko, Organizational Sciencies, Slovenia University of Baltimore, USA Sumy State University, Balatsky Academic and Scientific Institute of Franc Čuš, Finance, Economics and Management, CO-EDITORS / SOUREDNIKI University of Maribor, Slovenia Ukraine Petr Doucek Sasha M. Dekleva Hans Puxbaum, Prague University of Economics, DePaul University, School of Vienna University of Technology, Faculty of Informatics and Statistics, Accountancy and MIS, Chichago, USA Austria Czech Republic Vlado Dimovski, Vladislav Rajkovič, Matjaž Maletič University of Ljubljana, Slovenia University of Maribor, Slovenia University of Maribor, Faculty of Organizational Sciencies, Slovenia Daniel C. Ganster, Gábor Rekettye, Colorado State University, USA University of Pécs, Hungary Jože Zupančič University of Maribor, Faculty of Jože Gričar, Henk G. Sol, Organizational Sciencies, Slovenia University of Maribor, Slovenia Faculy of Economics and Business, University of Groningen, Netherlands Włodzimierz Sroka Werner Jammernegg WSB University, Department of Viena University of Economics and Eugene Semenkin Management, Dąbrowa Górnicza, Business Administration, Austria Reshetnev Siberian State University of Poland Science and Technology, Krasnoyarsk, Marius Alexander Janson, Russian Federation University of Missouri-St. Louis, USA Velimir Srića, Stefan Klein, University of Zagreb, Croatia University of Münster, Germany Paula Swatman, Aleksandar Marković, University of Tasmania, Australia University of Belgrade, Serbia Brian Timney, Hermann Maurer, Western University, Graz University of Technology, Graz, London, Ontario, Canada Austria Maurice Yolles, Matjaž Mulej, Liverpool John Moores University, UK University of Maribor, Slovenia Douglas R. Vogel, Valentinas Navickas, City University of Hong Kong, China Kaunas University of Technology, Lithuania Gerhard Wilhelm Weber, Poznan University of Technology, Ota Novotny, Poland University of Economics, Prague, Czech Republic Anna Lucyna Wziątek-Staśko, Jagiellonian University in Kraków, Milan Pagon, Poland Independent University, Bangladesh (IUB), Dhaka, Bangladesh Yvonne Ziegler, Frankfurt University of Applied Björn Paape, Sciences, Germany RWTH-Technical University Aachen, Germany Hans-Dieter Zimmermann, Eastern Switzerland University of Matjaž Perc Applied Sciences (OST), St. Gallen, University of Maribor, Slovenia Switzerland Dušan Petrač, NASA, Jet Propulsion Laboratory, California Institute of Technology, USA 318 Organizacija, Volume 57 Research Papers Issue 4, November 2024 DOI: 10.2478/orga-2024-0023 An Examination of Generational Differences in the Sharing Economy: Understanding the Motivations of Idle Asset Owners and Their Impact on Peer-to-Peer Platform Management Pavel PELECH1, Jaroslava DĚDKOVÁ2 1 Technical University of Liberec, Department of Marketing, Czech Republic 2 Technical University of Liberec, Faculty of Economics Technical, Czech Republic Background/Purpose: This research examines the dynamics of sharing idle assets on peer-to-peer (P2P) plat- forms within the sharing economy, focusing on understanding the motivational factors that influence asset owners across generations. Platforms that know what motivates asset owners to enter the sharing economy can more effec- tively set up and manage marketing communications and gain a competitive advantage. Methods: A mixed-methods approach has been adopted, which includes both quantitative and qualitative data collection and analysis. The analytical tools used include descriptive statistics, Kruskal-Wallis test, PCA and Welch ANOVA. Results: We identified four motivation factors. We found out that supply-side generational marketing is appropriate to use if the marketing message targets opportunistic factors (completely new) or social factors; for user-related and social factors, generational marketing is unnecessary, i.e., P2P platforms do not need to distinguish the segment they are targeting with the marketing communication. Conclusion: This research reveals how generational differences impact motivations to participate in the sharing economy, aiding P2P platform management. It offers a novel, comparative analysis of generational motivators, enriching sharing economy literature and providing practical insights for targeting different generations effectively. Understanding the motivations of idle asset owners is key to managing the business of P2P platforms. Keywords: Consumer behaviour, Generation, Management, Marketing, P2P, Sharing economy JEL Classification: D16, E21, M31 1 Introduction person who needs them at a certain time and under cer- tain pre-agreed conditions. In the P2P model, a platform mediates the sharing between these two entities. The plat- The essence of the sharing economy is the sharing and form operates in a sharing economy environment, which exploitation of idle assets. This economic model assumes is typically based on making a profit from brokering trans- that an asset owner shares his or her idle assets with a actions between the supply side and the demand side. It 1 Received: 21st November 2023; Accepted: 9th July 2024 319 Organizacija, Volume 57 Research Papers Issue 4, November 2024 must therefore know exactly who constitutes the supply information will provide a solid foundation for researchers side, who constitutes the demand side and how to reach and platforms in the sharing economy to understand what these two parties with its marketing communication. This motivates each generation to enter the sharing economy process can strengthen P2P competitiveness and help to from perspective of idle asset owners (supply side). Based gain a competitive advantage over other platforms and tra- on this, platforms can better set up a segmentation strategy ditional businesses. The more the sharing economy grows, and more effective marketing communications through a the more sustainability will be boosted. generational approach. Our research is based on a detailed For many years, companies offering their products literature review and quantitative and qualitative question- have used demographic and psychographic variables (Kot- ing. The main objective is to determine whether genera- ler & Armstrong, 2010, age segmentation, or a combina- tional marketing can be used to increase the size of the tion of them to segment their customers (Chaney et al., sharing economy and thus boost sustainability. 2017). Chaney et al. (2017) state that each age group has specific behaviours that companies can use to segment and set a specific marketing strategy. Although age is a frequent 2 Literature Review criterion for customer segmentation, generational segmen- tation may be more appropriate for companies to identify 2.1 Motivations for Entering the Sharing consumer motivations stemming from shared values and Economy beliefs across generations (Khare et al., 2012; Schewe & Meredith, 2004). Eastman & Liu (2012) even suggest that The motives for entering the sharing economy have consumption only depends on an individual’s generation been addressed by several authors, each of whom has affiliation, without demographic factors (gender, income, looked at the issue from a different perspective. The most education) playing any role. By using age segmentation important motive for the functioning of the sharing econ- in their marketing strategies, companies can enhance their omy is the willingness to share (Guyader et al., 2023). In competitiveness and achieve a competitive edge over other general, motivations can be divided into intrinsic motiva- companies that do not employ this approach. tors and extrinsic motivators (Hamari et al.,2015), which One way how companies can address the competitive- can bring both financial and non-financial benefits (Gazzo- ness issue is by understanding the generational cohorts. la et al., 2018). The importance of gaining competitive advantage in this An examination of internal and external factors was text lies in the fact that generational cohorts are not static conducted by Hamari et al. (2015). Internal factors include groups, but rather dynamic and evolving ones. As new gen- the pleasure of sharing and the possibility to promote sus- erations enter the market, they bring with them new pref- tainability, while external factors include gaining a good erences, expectations, and behaviours that may differ from reputation and economic reasons (saving money, easy ac- those of previous generations. We refer to the so-called cess to resources and the possibility to freely dispose of the generational cohort theory, in which individuals who ex- funds raised). Internal factors are considered stronger for perience similar historical, social, cultural, political, and participation in the sharing economy. They did not explore economic events between their 17th and 23rd year of life the importance of these factors in terms of the respond- are similar in their beliefs about fundamental social val- ents’ generational affiliation. A similar analysis was con- ues and behaviour (Mannheim, 1952). Companies should ducted by Guyader et al (2023). They investigated what identify the most substantial generational cohorts typical factors influence the population to share. Specifically, they of a company’s specific products and adjust the marketing looked at trend orientation, sharing orientation, citizen en- mix accordingly to these generations (Chaney et al., 2017). gagement, and authenticity of sharing within the platform. Different generations not only require different approach- They found that none of these factors had a direct effect on es within marketing, but also within management (Molek participation in the sharing economy (here specifically car et al., 2023). The sharing economy will follow this trend sharing). However, they mention that age may play a role, as well. Platforms must correctly define the customer seg- i.e. respondents of different ages will want to participate in ment (supply and demand side) they want to target with the sharing economy in different ways. Although Bäro et their marketing communications to remain competitive in al. (2022) looked at sociodemographic characteristics (sex, a rapidly changing market. By doing so, they can create age and education) that influence participation in the shar- value propositions that resonate with their target segments ing economy, but their research was only in general terms, and differentiate themselves from their competitors (Al- with no link to motivational factors. Moeller & Wittkowski jukhadar & Senecal, 2011). (2010) addressed internal factors. They found that if a per- To the best of our knowledge, there needs to be more son feels that the sharing follows a trend, they are more research examining the impact of generational cohorts on likely to participate in the sharing economy. Participants participants’ motivation to enter the sharing economy. We in the sharing economy must trust the platform through focused on this sustainability issue in our research. This which the sharing takes place (Lazakidou et al., 2008). It is 320 Organizacija, Volume 57 Research Papers Issue 4, November 2024 trust that, according to Martínez-González (2021), stands (Guyader et al., 20232018), meeting new people, socializ- at the interface between internal and external factors. ing (Gazzola et al., 2018), opportunity to be part of a com- We can also look at motivations regarding whether munity (Parameswaran & Whinston, 2007), interactivity, they are financial or non-financial. The financial benefits authenticity, warranty, legal protection, strict operating are entirely unquestionable. However, they have not been procedures (Fang & Li, 2022), grassroots engagement investigated separately in specific research but have been (Guyader et al., 2023), contributing to a significant change mentioned as other factors related to motivation for joining in the economy, flexibility of “doing business”, reduction the sharing economy. Fang & Li (2022) found that mone- of risk and liability arising from ownership (Benoit et al., tary savings, sustainability, interactivity, and authenticity 2017), or safe “business” (Yang & Ahn, 2016). motivate people to participate in the sharing economy. In- terestingly, they mention that it is also essential for these 2.2 Research Objective people that the platform provides sufficient guarantees to its users and legal protection and has set strict operational procedures. Gazzola et al. (2018) also came up with sim- Based on literature review, we ask several research ilar motivations. They found that sustainability, sociali- questions and derive specific hypotheses: zation, or economic factors motivate participants to par- 1. RQ1: What motivations are important for the idle ticipate in the sharing economy. In contrast, motivations asset owners? in the form of availability of idle assets and practicality 2. RQ2: Are motivations dependent on the genera- were found to be insignificant. Asset owners are eventu- tion of respondents? ally transformed into business entities, i.e. the financial • H0a: There is no significant relationship be- motivation is completely unquestionable (Mazurek, 2020). tween generations and their motivation. Among the non-financial benefits of the sharing econo- • H1a: There is significant relationship between my, authors often mention sustainability, as in Fang & Li generations and their motivation. (2022) and Gazzola et al. (2018). Sustainability emerges 3. RQ3: Is it possible to reduce the motivations as a future development direction in consumers’ consump- into some motivational factors? Are these fators tion decisions (Prothero et al., 2011). The more the popu- dependent on the generation of respondents? lation is part of the sharing economy, the better sustainable • H0a: There is no significant relationship be development can be achieved. Platforms need to realize tween generations and their motivation factors. that they can be instrumental in achieving this goal (Tu • H1a: There is significant relationship between et al., 2023). If platforms know what motivates owners of generations and their motivation factors. idle assets to enter the sharing economy, they can gain a 4. RQ4: Are there other motivations that affect idle competitive advantage and support the boost of the sharing asset owners in the sharing economy? economy. The first research question is related to the fact that The most extensive scholarly work to date regarding previous research has tended to look at the demand side of the factors affecting motivation was conducted by Ross- idle assets (Zhu et al., 2017). In contrast, research exam- mannek & Chen (2023). They summarized the results of ining the supply side has been relatively sparse (Rossman- 152 studies. They found that the authors of these studies nek & Chen, 2023; Bucher et al., 2016). One of the few identified a total of 26 motivations (altruism, attitude, de- studies that distinguish between the supply and demand sire for uniqueness, economic benefit/motive, emotional sides suggests a mismatch between the motivations of us- value, enjoyment, familiarity, hedonic value/motivation, ers of idle assets and their owners to use sharing economy information quality, materialism, perceived behavioural services (Bellotti et al., 2015). However, more is needed to control, perceived ease of use, perceived risk, perceived understand the issue in its full complexity, as in this study, usefulness/value/benefits, price, reputation, satisfaction, the authors examined only a few motivations (ideological service quality, sharing intention, social benefits, social motivation, community, supply-side sustainability, need influence, subjective norm, sustainability value/motiva- satisfaction, value enhancement, and demand-side con- tion, technical quality, trust, utilitarian motivation/bene- venience). In the study by Böcker & Meelen (2017), we fits, word of mouth). They also analysed these motivations find three more motivations (economic, environmental, concerning the influence of the sharing sector (ride sharing, and social), which the authors examined from a supply and home sharing, fashion sharing, and car sharing), location demand perspective concerning the object of the sharing (China, Europe, South/Southeast Asia, and the USA), and economy (accommodation, car, tool, ride, meal sharing). whether it is supply or demand. In this summary, they have Benoit et al. (2017) distinguish the motives of the differ- analysed articles from 2015 to 2022. Given our analysis, ent actors. A customer who needs to share a product with this paper cannot be considered exhaustive. We found an- someone else is motivated by financial, social, and utility other motivation factors discussed by other authors: trend factors and wants to reduce the risk and liability of own- orientation (Guyader et al., 2023), proenvironmentalism ership by sharing products. The owner of the asset is mo- 321 Organizacija, Volume 57 Research Papers Issue 4, November 2024 tivated by similar factors. The financial and social factors important for Generation Y, i.e., they are a more socially, are the same, but another factor is referred to as business culturally, and environmentally conscious generation than flexibility. As for the platform itself, it is also motivated other generations. This opinion is confirmed by Jelinkova by financial factors. However, the big motivation is that it et al. (2021). They state that the younger a person is, the creates a relationship between supply and demand, and at more important the sharing economy’s economic, social, the same time, it is involved in a significant change in the and environmental aspects are for them. Of note is the re- economy, which the sharing economy undoubtedly is. In- search of Martínez-González (2021), who examined nine teresting research is presented by Angelovska et al. (2020). motivations for Generation Z. They concluded that attitude They investigate what factors motivate sharing economy and social norms are important motivators for this genera- customers to become providers. They find that custom- tion. According to them, general trust in the sharing econo- ers motivated by altruistic motives and less by financial my links intrinsic and extrinsic motivators and encourages motives are more likely to become providers, i.e., to cre- this generation to participate in the sharing economy. We ate a supply side in the sharing economy. Given the more are unaware of other scholarly publications on the issue of than three dozen motivations identified, further research is comparing Generations X, Y, and Z, which opens up space needed to continue and expand the knowledge platforms for further research. in this area. The fourth research question is related to the extensive The second and third research question is related to the study by Rossmannek & Chen (2023). It has been shown fact that previous research has primarily not addressed the that there are other motivations elaborated on by the au- generational perspective. Authors who have addressed the thors in their papers, but the work of these two authors generational perspective in conjunction with motivations does not mention them. Therefore, the question arises as for participation in the sharing economy are much fewer to whether other motivations have not been considered by than those who have addressed motivators from a gener- previous research that are important for the functioning of al population perspective. For example, the generational platforms and supporting sustainability. perspective in terms of motivations for participation in the sharing economy has been explored by Alemi et al. (2018). They found that members of Generation Y are more likely 3 Methodology to use Uber/Lyft services. In addition to the generation- al influence on this outcome, they report that those who 3.1 Questionnaire Development travel, use smartphones more often, are pro-environment, tech-oriented, and desire diversity are more likely to par- Based on a search of professional publications and re- ticipate in the sharing economy. These last three factors search, we have compiled the first version of the question- are additional motivating factors influencing participation naire for our research investigation. We ensured content in the sharing economy. Motivations (attitude, subjective validity. The content validity of the data was rigorously norm, and perceived behavioural control) in Generation evaluated through consultation with two subject matter ex- Z were investigated by Pham et al. (2021). They found perts in the field. All content validity indices were above that perceived behavioural control significantly influenc- the minimum value of 0.8 (Yusoff et al., 2021), i. e., the es whether this generation will demand the sharing mode. questionnaire´s scale has achieved a satisfactory level of Intergenerational research was also noted by Mahadevan content validity. We collected data in two phases. (2018), who looked at the motives of Baby Boomers, X The first phase was qualitative data collection, for and Y generations, among others, and compared how these which we used exploratory research using the focus group generations differed from each other on selected motiva- method. This research took place in early November 2022. tions. He found, for example, that the difference between Nine people participated in this research, evenly distrib- Generation X and Y is statistically significant for a motiva- uted among the generations of respondents X, Y, and Z tion that is generally related to the sharing economy philos- analysed. In particular, the qualitative research revealed ophy. However, he only looked at the demand side of the that in addition to general motivators, respondents might sharing accommodation. Generation Y was also addressed be motivated by platform-related motivators and idle as- by Ranzini et al. (2017). They defined four motivations for sets. Based on the exploratory research, we enriched the entering the sharing economy, which they called econom- first version of the questionnaire with additional motiva- ic, convenience, fun/social interaction, and reciprocity fac- tions and finalized the questionnaire for quantitative data tors. They included cost reduction and additional income collection. in the economic motives. We can find further research fo- We divided the questionnaire into five parts. Due to the cusing on Generation Y in Sahelices-Pinto et al. (2021) or large amount of information collected, we only evaluated Lee et al. (2019). The authors mentioned above focused on data related to motivations for this paper from the first and the issue of sustainability as a motivating factor for Gen- fifth part of questionnaire. First, we asked respondents if eration Y. They concluded that environmental issues are they knew what the sharing economy was. If they did not 322 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 1: Motivations Overview. Side Abbreviation Full Name O_1 I will make some extra money O_2 I will meet new people O_3 I will help those who cannot afford the item by sharing O_4 I will rise with people around me (image) by sharing O_5 I support sustainability Supply O_6 The app or website is user-friendly O_7 The sharing platform is known O_8 The platform can deal with insurance or damage to the loaned item O_9 The platform transparently handles user peer reviews O_10 I will be favoured if I ever want to borrow something myself O_11 I have trustworthy information about the customer Source: Own know, they could read a brief explanation of what can be characteristic (Johnson, 2014). The age limit for partici- understood by the sharing economy. The first part of the pation in this research was 18 years, and access to the in- questionnaire focused on the motivating factors from the ternet was a prerequisite for participation in the sharing perspective of the idle asset owner and the customer in- economy. A total of 963 respondents participated in the terested in sharing the asset. The fifth part included demo- research. We excluded 90 respondents because they were graphic data (gender, birth year, residence size, education- older than our research analysed. Finally, we have left 873 al attainment, and region). Table 1 provides an overview of completely filled out questionnaires of the study popula- these analysed motivations. tion of respondents. Respondents could express how motivated they would To compare the sample of respondents with the total be by the analysed attributes on the following ordinal population, we used data from the Czech Statistical Office scale: (1) definitely not; (2) rather not; (4) rather yes; (5) (Census, 2021) and information collected by the Ministry definitely yes. We did not deliberately offer the middle op- of the Interior of the Czech Republic. In the Czech Repub- tion (3) to respondents, as we are inclined to the view of lic in 2021, 51.42% of men and 48.58% of women lived in Sturgis et al. (2012), who states that when a respondent is the analysed population of respondents (Generation X, Y, given a choice of a middle answer, the researcher creates Z). Of these persons, Generation X accounted for 40.04%, space for the respondent to avoid answering the question. Generation Y 33.94%, and Generation Z 26.02%. In our Lucian (2016) argues that neutral attitudes are not pos- sample, we had 31.2% Generation X respondents, 31.5% sible as the respondent tends towards a specific answer. Generation Y respondents, and 37.3% Generation Z re- Although Adelson & McCoach (2010) state that midpoint spondents. Information from the Ministry of the Interior surveys are more reliable, Nadler et al. (2015) reject this, of the Czech Republic shows that as of 1 January 2023, stating that midpoint and non-centerpoint surveys generate 52.42% of the population over 15 years of age in the Czech similar results. Republic lived in places with populations under 10,000 inhabitants, 20.68% lived in places with populations un- 3.2 Data Collection and Sample Profile der 50,000, 7.99% lived in places with populations under 100,000, and 19.91% lived in places with populations over 100,000. Our sample of respondents was over 18 years old The second phase involved collecting quantitative data and limited by Generation X, i.e., we cannot make an ex- using an online survey. We collected data from December act comparison with the population of the Czech Repub- 2022 to April 2023 in the Czech Republic through an on- lic. The structure shows that our sample is very similar line Google form. We used snowball sampling to obtain to the population of the Czech Republic. To compare the responses from respondents of different generations who educational background of the population, we again used are difficult to reach. We chose the snowball method to data from the Czech Statistical Office (Census, 2021). We obtain a reasonably representative sample of respondents focused on the 18+ population with the age restriction of (Kirchherr & Charles, 2018). It is based on referring ini- Generation X. We found that in the Czech Republic in tially selected respondents to others who have a particular 2021, a total of 7.91% of people over 18 years of age lived 323 Organizacija, Volume 57 Research Papers Issue 4, November 2024 with primary education, 66.51% of the population with null hypothesis that the data is not normally distributed. secondary education, 2.38% with higher vocational edu- In the second phase, we described the supply side mo- cation and 23.20% with university education. Our sample tivations using descriptive statistics (mean, SD). Given of respondents reflected this structure similarly. As in the the large number of supply side motivations, we decided population, the largest proportion of our sample consisted to evaluate the data using principal component analysis of respondents with a higher professional education, fol- (PCA) and varimax rotation method in order to identify lowed by respondents with a university education. the main motivating factors for idle asset owners to enter Since our sample corresponds to the population distri- the sharing economy. Differences between generations of bution in the Czech Republic, it is replicable in terms of in- respondents in their primary motivation were evaluated ternational comparisons while maintaining the population using the Kruskal-Wallis test. Differences in factor load- distribution in the selected country and the population can ings between generations of respondents were evaluated be compared in research by other authors. using the Welch ANOVA. 3.3 Statistical Methods 4 Results We performed data analysis using SPSS statistical 4.1 RQ1: Important Motivations for Idle software. First, we analysed data reliability. For the data Asset Owners analysed (11 items), we checked the reliability using Cronbach’s alpha. The value of this indicator was 0.811, indicating high reliability. Hair et al. (2013) state that it We investigated an answer to RQ1 using descriptive In is important that the value exceeds the threshold of 0.7 in Table 2 we can see the differences between the three ana- order for the data to be considered reliable. We performed lysed generations based on descriptive statistics. From Ta- a normality test on these 11 items. Kolmogorov-Smirnov ble 2, we see that the strongest motivation for asset owners and Shapiro-Wilk showed significance of less than 0.001 is O_11 for Generation X. On the supply side, the strongest for all items, which is less than the alpha value of 0.05. motivation for Generation Y and Z is O_1. On the other These results are reason to think that our data differs sig- hand, the weakest motivation is O_4 for all generations nificantly from a normal distribution. This may be due to a analysed, consistently. missing middle value of the answer. We cannot reject the Table 2: Descriptive Statistics – Supply Side Mean SD Motivations X Y Z X Y Z O_1 3.93 4.16 4.52 1.241 1.080 0.833 O_2 2.68 2.56 2.77 1.357 1.276 1.321 O_3 3.56 3.63 3.96 1.296 1.215 1.066 O_4 2.21 2.02 2.47 1.260 1.194 1.333 O_5 3.60 3.57 3.86 1.259 1.333 1.179 O_6 3.48 3.26 3.57 1.291 1.346 1.218 O_7 3.50 3.52 3.58 1.260 1.236 1.240 O_8 3.82 4.08 4.09 1.294 1.159 1.092 O_9 3.70 3.80 3.84 1.307 1.175 1.126 O_10 3.67 3.78 3.93 1.314 1.236 1.120 O_11 3.95 4.10 4.08 1.286 1.088 1.168 Source: Own 324 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 3: Kruskal-Wallis test – Supply Side Multiple Comparisons Mot. K-W H Asymp. Sig. (p-value) Sig. ε2 - adjusted significance (ε2) X:Y X:Z Y:Z O_1 49.917 <0.001 S 0.057 0.097 <0.001 (0.079) <0.001 (0.007) O_2 4.788 0.091 NS x x x x O_3 18.132 <0.001 S 0.021 1.000 <0.001 (0.022) <0.001 (0.022) O_4 21.446 <0.001 S 0.023 0.122 <0.001 (0.010) 0.041 O_5 10.345 0.006 S 0.012 1.000 0.018 0.018 O_6 9.230 0.010 S 0.011 0.088 0.010 (0.001) 1.000 O_7 0.820 0.664 NS x x x x O_8 6.433 0.040 S 0.007 0.097 0.068 1.000 O_9 0.268 0.875 NS x x x x O_10 4.909 0.086 NS x x x x O_11 0.910 0.634 NS x x x x Source: Own 4.2 RQ2: Dependence of Motivations on Generation Y, 448.41 for Generation Z). Our data provide Generation strong evidence against the null hypothesis only for some motivators. The p-value is the probability of obtaining an We used the Kruskal-Wallis test to compare the me- effect at least as extreme as the one in our sample, assum- dians of three groups of respondents. If the value of H is ing the null hypothesis is true. The smaller the p-value, the greater than the critical chi-square distribution, we can less likely it is that we would observe the effect if the null reject the null hypothesis and accept that at least one hypothesis were true, and the more likely it is to reject the group is different from the others. Table 3 shows that the null hypothesis. Based on our results, we reject the null Kruskal-Wallis test showed statistically significant differ- hypothesis H0a and accept the alternative hypothesis H1a ences on the significance level 0.05 for some motivators for O_1, O_3, O_4, O_5, O_6 and O_8. The p-value is less on the supply side. than 0.05, indicating that the observed effect is statistically The Kruskal-Wallis H test indicated that there is a sig- significant for these motivators. There were no significant nificant difference in the dependent variable between the differences by motivators O_2, O_7, O_9 and O_10. Our different groups for O_1 (χ2(2) = 49.917, p < 0.001, with research showed that platforms can adopt different market- a mean rank score of 375.98 for Generation X, 417.66 ing strategies based on the type of motivators. for Generation Y, 504.23 for Generation Z), O_3 (χ2(2) For pairwise comparisons we used the Post-Hoc = 18.132, p < 0.001, with a mean rank score of 409.18 for Dunn’s test using a Bonferroni adjusted alpha (adjust- Generation X, 412.91 for Generation Y, 480.53 for Gen- ed α = 0.01667). As we can see from the Table 3, after eration Z), O_4 (χ2(2) = 21.446, p < 0.001, with a mean adjusting for multiple comparisons using the Bonferroni rank score of 432.11 for Generation X, 390.57 for Gener- correction method, the result was not statistically signif- ation Y, 480.24 for Generation Z), O_5 (χ2(2) = 10.345, icant for all groups of respondents. We also calculated ε2 p = 0.006, with a mean rank score of 417.24 for Genera- (epsilon-squared) to verify effect size of our results. This tion X, 417.28 for Generation Y, 470.12 for Generation Z), coefficient does not have any strict intervals, so it is dif- O_6 (χ2(2) = 9.230, p = 0.010, with a mean rank score of ficult to make clear conclusions. Mangiafico (2016) ex- 446.05 for Generation X, 402.19 for Generation Y, 458.81 plained these values as follows: 0.01-<0.08 (small effect), for Generation Z), and O_8 (χ2(2) = 6.433, p = 0.040, with 0.08-<0.26 (medium effect), above 0,26 = large effect). a mean rank score of 407.27 for Generation X, 452.89 for López-Martín & Ardura-Martínez (2023) set up different 325 Organizacija, Volume 57 Research Papers Issue 4, November 2024 intervals: <0.01 (very small effect), 0.01<0.05 (small ef- in 8 iterations. fect), 0.06<0.13 (moderate effect), >0.14 (large effect), Table 4 shows that the first component consists of mo- 0.36<0.64 (strong effect), 0.64<1 (very strong effect). As tivations O_10, O_11, O_8, O_1, O_9. These are motiva- Lakens (2013) emphasized, even small effect sizes can tions that are somehow related to the financial and non-fi- have big impacts. nancial benefits that the owner of the idle asset can gain by entering the sharing economy. Hence, this factor has been 4.3 RQ3: Reducing Motivators into termed as opportunistic. The second component consists Motivating Factors of the motivations O_6 and O_7. These include motiva- tions that are related to the platform. For this reason, this factor was named as user-related. The third component The literature review and our own research showed consists of motivations O_4 and O_2. These include mo- that there are a number of motivations for entering the tivations that are related to people. For this reason, this sharing economy. Platforms that need to target specific factor was named as social. The fourth component consists customer segments with their marketing communications of motivations O_5 and O_3. These include motivations need to have information on what motivates which cus- that are related to society. For this reason, this factor was tomer segment. We investigated an answer to RQ3 using named as societal. PCA with varimax rotation. The KMO value was 0.847 To target marketing communications, platforms need and Bartlett’s Test of Sphericity was significant (<0.001), to know if factors are relevant to all generations. Factor indicating the analysis is appropriate. Initial eigenvalues loadings (regression factor score) for each respondent exceeded 1.0 for two components, and were very close to were used to assess statistical significance and test the this value for the other two components. For this reason, validity of hypothesis H0b. We used Welch ANOVA. We 4 components were selected for extraction. All commu- also calculated η2 (eta-squared) to verify effect size of our nalities were greater than 0.4. We used PCA with varimax results. We found out interpretation by Richardson (2011). rotation and Kaiser normalization. In Table 4 we see the He set up these intervals: small (η2 = 0.01), medium (η2 rotated component matrix. We used extraction method = 0.06), and large (η2 = 0.14) effects. As Lakens (2013) Principal Component Analysis and rotation method vari- emphasized, even small effect sizes can have big impacts. max with Kaiser normalization. Rotation was converged The results are captured in Table 5. Table 4: Rotated Component Matrix – Supply Side Motivations Component 1 Component 2 Component 3 Component 4 (Opportunistic) (User-related) (Social) (Societal) O_10 0.738 O_11 0.712 O_8 0.650 0.434 O_1 0.639 0.409 O_9 0.557 0.440 O_6 0.828 O_7 0.788 O_4 0.817 O_2 0.720 0.342 O_5 0.774 O_3 0.746 Source: Own 326 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 5: Welch ANOVA – Supply Side Multiple Comparisons (adjus- Factor F P-value Significance η2 ted p-value) X:Y X:Z Y:Z Opportunistic 9.749 <0.001 S 0.024 0.002 <0.001 NS User-related 1.309 0.271 NS x x x x Social 12.221 <0.001 S 0.027 NS 0.003 <0.001 Societal 1.590 0.205 NS x x x x Source: Own Table 5 shows strong evidence against the null hy- from the traditional factors. In fact, none of the authors that pothesis. A higher F value means significant differences we reviewed in our systematic literature review mentioned between group means. The p-value is less than 0.05, in- this factor or its implications for sharing economy motiva- dicating a statistically significant effect. The smaller the tion. This new factor consists of motivations O_10, O_11, p-value, the more likely it is to reject the null hypothesis. O_8, O_1, O_9. These are motivations that are somehow Based on our results, we reject the null hypothesis H0b and related to the financial and non-financial benefits that the accept the alternative hypothesis H1b for the factor named owner of the idle asset can gain by entering the sharing “opportunistic” (F (2, 870) = 9.749, p <0.001) and “social” economy. (F (2, 870) = 12.211, p <0.001). For platforms, this means that supply-side generational marketing is appropriate to 4.5 Results Comparison with Other use if the marketing message targets opportunistic factors or social factors. Post hoc comparisons using Bonferroni International Results indicated by opportunistic factor that the regression factor scores differ between generations X (M = -0.2245, SD = One of the benefits of generational segmentation is 1.1060) and Y (M = 0.0639, SD = 1.0203), lower bound that it can help companies gain a competitive advantage of 95 % confidence interval = -0.4913, upper bound of 95 in the market by understanding the needs and preferenc- % confidence interval = -0.0855; and X (M = - 0.2245, es of different generations and offering them products and SD = 1.1060) and Z (M = 0.1336, SD = 0.8497), lower services that match their values and lifestyles (Eastman & bound of 95 % confidence interval = -0.5527, upper bound Liu, 2012). For example, millennials are more likely to of 95 % confidence interval = -0.1630. Post hoc compar- value social responsibility, environmental sustainability, isons using Bonferroni indicated by social factor that the and digital convenience than older generations (Chaney regression factor scores differ between generations X (M = et al., 2017). Therefore, companies that can communicate -0.0635, SD = 0.9784) and Z (M = 0.2060, SD = 0.9848), their social and environmental impact, as well as provide lower bound of 95 % confidence interval = -0.4640, upper online and mobile platforms for their customers, may have bound of 95 % confidence interval = -0.0750; and Y (M = an edge over their competitors in attracting and retaining - 0.1813, SD = 0.9990) and Z (M = 0.2060, SD = 0.9848), millennials. P2P platforms that can demonstrate their ser- lower bound of 95 % confidence interval = -0.5812, upper vice excellence, offer personalized communications may bound of 95 % confidence interval = -0.1934. In contrast, have an advantage over their competitors. By using gener- for user-related and social factors, generational marketing ational segmentation, companies can create more effective is unnecessary, i.e. platforms do not need to distinguish the marketing strategies that appeal to the specific character- segment they are targeting with the marketing communi- istics of each generation and gain a competitive advantage cation in these cases. in the market. Our research focused on three generations: X (born between 1965 and 1980), Y (born between 1981 4.4 RQ4: New Motivation Factor and 1996), and Z (born after 1997) explained by Dimock (2022). These generations have different characteristics, experiences, and expectations that may influence their atti- The answer to RQ4 is closely related to the results of tudes and behaviours toward the sharing economy. the PCA analysis. We found that the new factor can be Our first research question (RQ1) aimed to find out what considered as opportunistic factor. This factor is different motivations are important on the supply and demand side 327 Organizacija, Volume 57 Research Papers Issue 4, November 2024 in generation perspective. In other international research social responsibility, environmental sustainability, and articles was difficult to find information about motivations digital convenience than older generations (Chaney et al., on the supply side. Research examining the supply side 2017). Therefore, companies that can communicate their has been relatively sparse (Rossmannek & Chen, 2023). social and environmental impact, as well as provide on- As a result, supply-side motivations remain an unexplored line and mobile platforms for their customers, may have area (Bucher et al., 2016). Also, studies focusing on supply an edge over their competitors in attracting and retaining and demand side together have been relatively sparse. One millennials. P2P platforms that can demonstrate their ser- of the few studies that distinguish between the supply and vice excellence, offer personalized communications may demand sides suggests a mismatch between the motiva- have an advantage over their competitors. By using gener- tions of users of idle assets and their owners to use sharing ational segmentation, companies can create more effective economy services (Bellotti et al., 2015). In the study by marketing strategies that appeal to the specific character- Böcker & Meelen (2017), we find three more motivations istics of each generation and gain a competitive advantage (economic, environmental, and social), which the authors in the market. Our research focused on three generations: examined from a supply and demand perspective concern- X (born between 1965 and 1980), Y (born between 1981 ing the object of the sharing economy (accommodation, and 1996), and Z (born after 1997) explained by Dimock car, tool, ride, meal sharing). We didn´t find any generation (2022). These generations have different characteristics, context in research articles. Further research was needed to experiences, and expectations that may influence their atti- continue and expand the knowledge platforms in this area. tudes and behaviours toward the sharing economy. In the second research question (RQ2) and third re- Our first research question (RQ1) aimed to find out what search question (RQ3) we aimed to find out if motivations motivations are important on the supply and demand side were dependent on the generation of respondents (RQ2) in generation perspective. In other international research and if it was possible to reduce the motivations into some articles was difficult to find information about motivations motivational factors, especially in generation perspective on the supply side. Research examining the supply side (RQ3). Other international research articles focused often has been relatively sparse (Rossmannek & Chen, 2023). only on one generation (Alemi et al., 2018; Pham et al., As a result, supply-side motivations remain an unexplored 2021; Ranzini et al., 2017; Sahelices-Pinto et al., 2021; area (Bucher et al., 2016). Also, studies focusing on supply Lee et al., 2019). We can partly compare our result with the and demand side together have been relatively sparse. One research study of Mahadevan (2018), who looked at the of the few studies that distinguish between the supply and motives of Baby Boomers, X and Y generations, among demand sides suggests a mismatch between the motiva- others, and compared how these generations differed from tions of users of idle assets and their owners to use sharing each other on selected motivations. We illustrated the im- economy services (Bellotti et al., 2015). In the study by portance of tailoring marketing communication in the shar- Böcker & Meelen (2017), we find three more motivations ing economy according to the generational cohorts and the (economic, environmental, and social), which the authors intended message for the supply side of the platform. examined from a supply and demand perspective concern- The last fourth research question (RQ4) tried to answer ing the object of the sharing economy (accommodation, if there were other motivation factors that had not been car, tool, ride, meal sharing). We didn´t find any generation mention before. We could consider “opportunistic factor” context in research articles. Further research was needed to as a new motivation factor in the sharing economy. This continue and expand the knowledge platforms in this area. factor included motivations that are somehow related to In the second research question (RQ2) and third re- the financial and non-financial benefits that the owner of search question (RQ3) we aimed to find out if motivations the idle asset can gain by entering the sharing economy. were dependent on the generation of respondents (RQ2) These motivations were clearly opportunistic. and if it was possible to reduce the motivations into some motivational factors, especially in generation perspective 5 Discussion and Conclusions (RQ3). Other international research articles focused often only on one generation (Alemi et al., 2018; Pham et al., 2021; Ranzini et al., 2017; Sahelices-Pinto et al., 2021; 5.1 Results Comparison with Other Lee et al., 2019). We can partly compare our result with the International Results research study of Mahadevan (2018), who looked at the motives of Baby Boomers, X and Y generations, among One of the benefits of generational segmentation is that others, and compared how these generations differed from it can help companies gain a competitive advantage in the each other on selected motivations. We illustrated the im- market by understanding the needs and preferences of dif- portance of tailoring marketing communication in the shar- ferent generations and offering them products and services ing economy according to the generational cohorts and the that match their values and lifestyles (Eastman &and Liu, intended message for the supply side of the platform. 2012). For example, millennials are more likely to value 328 Organizacija, Volume 57 Research Papers Issue 4, November 2024 The last fourth research question (RQ4) tried to answer be achieved by understanding and satisfying the diverse if there were other motivation factors that had not been needs and preferences of different generations of consum- mention before. We could consider “opportunistic factor” ers who participate in the sharing economy. By doing so, as a new motivation factor in the sharing economy. This the sharing economy platforms and providers can attract factor included motivations that are somehow related to and retain more customers, increase their market share and the financial and non-financial benefits that the owner of profitability, and enhance their reputation and social im- the idle asset can gain by entering the sharing economy. pact. This research can help them to identify the key factors These motivations were clearly opportunistic. that motivate and influence different generations to engage in the sharing economy, as well as their expectations and 5.2 Main Findings of the Research preferences for the products and services offered by the sharing economy platforms and providers. By comparing and contrasting the generational differences, this research The sharing economy can play a significant role in pro- can also help them to develop and implement more effec- moting sustainable development by facilitating the efficient tive and efficient strategies for segmenting, targeting, and use of resources and minimizing the environmental foot- positioning their offerings in the sharing economy market. print of consumption. By understanding the varied needs Furthermore, this research can help them to anticipate and and desires of consumers from different generations, shar- respond to the potential challenges and opportunities that ing economy platforms can adopt more sustainable busi- may arise from the changing generational dynamics in the ness practices that decrease waste, and increase resource sharing economy. For theory, it can contribute to the exist- efficiency. By integrating environmental considerations ing literature on the sharing economy by providing a com- into their business strategies, sharing economy platforms prehensive and comparative analysis of the generational can improve their reputation, appeal to eco-conscious con- differences in the motivators and preferences of sharing sumers, and secure a competitive edge in the marketplace. economy. The novelty of our research is that, to the best of our We found out that supply-side generational market- knowledge, no author has investigated the effect of three ing is appropriate to use if the marketing message targets generational cohorts on their motivation to participate in opportunistic factors or social factors. In contrast, for us- the sharing economy. We know that segmentation by gen- er-related and social factors, generational marketing is erational cohorts has been shown to be useful in a market unnecessary, i.e. platforms do not need to distinguish the economy (Fukuda, 2009). But we lack this confirmation segment they are targeting with the marketing communi- in the sharing economy. While in a market economy a cation in these cases. person buys a car with a long-term vision of using it, in We can see further contributions in that the research a sharing economy a person wants to use a shared car in addresses the supply side, assessing motivational factors the short term. Their motivations may differ. Our research in general, and motivational factors associated with the in the sharing economy environment is unique. We found platform. All this is in a generational context. The genera- that each generation has different supply-side motivators, tional perspective can help platforms understand the par- i.e. each generation may be motivated by a different mar- ticipants in the sharing economy and better tailor market- keting message. The essence of this research is not only ing communications to a specific generation, considering to increase the volume of the shared economy as a tool to what motivates that generation to join the sharing econo- increase sustainability, but also to reduce marketing costs my. Effective marketing creates value for the organization by precisely targeting a specific marketing message to a (Garbarski, 2014). Linking this information to other work specific generation in the sharing economy. Generational by the authors of this research, platforms gain additional marketing management can therefore be a valuable strate- information on what generation to reach and how to mo- gy to improve customer acquisition and retention not only tivate if they need to increase supply in the sharing econ- in the market economy, but also in the sharing economy. omy. Our study sought to put forward workable approach- es to help increase the size of the sharing economy. One of the main objectives of this research was to explore the 5.3 Limitations and Future Research importance of gaining competitive advantage in the shar- ing economy. Competitive advantage refers to the abili- This research has several limitations that point to areas ty of an organization to create and sustain superior value for further exploration of the motivators of sharing econ- for its customers and stakeholders relative to its competi- omy participation in a generational context. Firstly, the tors. For practice, it can provide valuable insights for the sample size was relatively small and restricted to a single sharing economy platforms and providers who want to country, which may limit the applicability of the results. target different generations more effectively and efficient- Future studies could use more extensive and varied sam- ly. P2P platforms can achieve competitive advantage can ples from multiple countries and cultures to investigate 329 Organizacija, Volume 57 Research Papers Issue 4, November 2024 cross-cultural motivational differences in sharing econ- Böcker, L., & Meelen, T. (2017). Sharing for people, omy. Secondly, the study used self-reported motivation planet or profit? analysing motivations for intended measures, which may be influenced by social desirability sharing economy participation. Environmental Inno- bias or memory errors. Future research could employ more vation and Societal Transitions, 23, 28–39. https://doi. objective and behavioural motivation measures to confirm org/10.1016/j.eist.2016.09.004. and supplement the self-reported data, such as actual usage Bucher, E., Fieseler, C., & Lutz, C. (2016). What’s mine data or experiments. Thirdly, while the study examined the is yours (for a nominal fee) – exploring the spectrum main effects of generational cohorts on sharing economy of utilitarian to altruistic motives for internet-mediated motivators, it did not consider potential moderating or me- sharing. Computers in Human Behaviour, 62, 316– diating factors that could account for generational differ- 326. https://doi.org/10.1016/j.chb.2016.04.002. ences. Future research could explore other factors, such as Census 2021. Obyvatelstvo | Sčítání 2021. Retrieved personality traits, values, attitudes, or situational variables, May 2, 2023, from https://vdb.czso.cz/vdbvo2/faces/ that might affect generational cohorts’ relationship with cs/index.jsf?page=statistiky&katalog=33495#kata- sharing economy motivations. log=33495. Dimock, M. (2022). Defining generations: Where mil- Literature lennials end and generation Z begins. Pew Research Center. Retrieved March 14, 2023, from https://www. pewresearch.org/fact-tank/2019/01/17/where-millen- Adelson, J. L., & McCoach, D. B. (2010). Measuring the nials-end-and-generation-z-begins/. mathematical attitudes of elementary students: The Eastman, J. K., & Liu, J. (2012). The impact of genera- effects of a 4-point or 5-point Likert-type scale. Ed- tional cohorts on status consumption: An exploratory ucational and Psychological Measurement, 70(5), look at generational cohort and demographics on status 796–807. https://doi.org/10.1177/0013164410366694. consumption. Journal of Consumer Marketing, 29(2), Alemi, F., Circella, G., Handy, S., & Mokhtarian, P. (2018). 93–102. https://doi.org/10.1108/07363761211206348. What influences travellers to use uber? exploring the Fang, Y.-H., & Li, C.-Y. (2022). Does the sharing economy factors affecting the adoption of on-demand ride ser- change conventional consumption modes? Internation- vices in California. Travel Behaviour and Society, 13, al Journal of Information Management, 67, 102552. 88–104. https://doi.org/10.1016/j.tbs.2018.06.002. https://doi.org/10.1016/j.ijinfomgt.2022.102552. Aljukhadar, M., & Senecal, S. (2011). Segmenting Fukuda, K. (2009). A cohort analysis of household vehicle the online consumer market. Marketing Intelli- expenditure in the U.S. and Japan: A possibility of gen- gence & Planning, 29(4), 421–435. https://doi. erational marketing. Marketing Letters, 21(1), 53–64. org/10.1108/02634501111138572. https://doi.org/10.1007/s11002-009-9077-2. Angelovska, J., Čeh Časni, A., & i Lutz, C. (2020). TURN- Garbarski, L. (2014). Creating Value for the Organization ING CONSUMERS INTO PROVIDERS IN THE through Marketing. Central European Management SHARING ECONOMY: EXPLORING THE IMPACT Journal, 22(4), 60-69. https://doi.org/10.7206/mba. OF DEMOGRAPHICS AND MOTIVES. Ekonomska ce.2084-3356.119. misao i praksa, 29 (1), 79-100. Retrieved from: https:// Gazzola, P., Vătămănescu, E.-M., Andrei, A. G., & Mar- hrcak.srce.hr/239585. rapodi, C. (2018). Users’ motivations to participate in Bäro, A., Toepler, F., Meynhardt, T., & Velamuri, V. K. the sharing economy: Moving from profits toward Sus- (2022). Participating in the sharing economy: The role tainable Development. Corporate Social Responsibil- of individual characteristics. Managerial and Decision ity and Environmental Management, 26(4), 741–751. Economics, 43(8), 3715–3735. https://doi.org/10.1002/ https://doi.org/10.1002/csr.1715. mde.3624. Guyader, H., Olsson, L. E., & Friman, M. (2023). Sharing Bellotti, V., Ambard, A., Turner, D., Gossmann, C., economy platforms as mainstream: Balancing pro-so- Demkova, K., & Carroll, J. M. (2015). A muddle of cial and economic tensions. Total Quality Management models of motivation for using peer-to-peer economy & Business Excellence, 1–20. https://doi.org/10.1080/ systems. Proceedings of the 33rd Annual ACM Confer- 14783363.2022.2159366. ence on Human Factors in Computing Systems. https:// Hair, J. F., Ringle, C. M., & Sarstedt, M. (2013). Partial doi.org/10.1145/2702123.2702272. least squares structural equation modeling: Rigorous Benoit, S., Baker, T. L., Bolton, R. N., Gruber, T., & Kan- applications, better results and higher acceptance. dampully, J. (2017). A triadic framework for collabo- Long Range Planning, 46(1-2), 1–12. https://doi. rative consumption (CC): Motives, activities and Re- org/10.1016/j.lrp.2013.01.001. sources & capabilities of actors. Journal of Business Hamari, J., Sjöklint, M., & Ukkonen, A. (2015). The shar- Research, 79, 219–227. https://doi.org/10.1016/j.jbus- ing economy: Why people participate in collaborative res.2017.05.004. consumption. Journal of the Association for Infor- 330 Organizacija, Volume 57 Research Papers Issue 4, November 2024 mation Science and Technology, 67(9), 2047–2059. Mannheim, K. (1952). The sociological problem of gen- https://doi.org/10.1002/asi.23552. erations. Essays on the Sociology of Knowledge, 306, Chaney, D., Touzani, M., & Ben Slimane, K. (2017). Mar- 163-195. keting to the (new) generations: Summary and Per- Martínez-González, J. A., Parra-López, E., & Barrientos- spectives. Journal of Strategic Marketing, 25(3), 179– -Báez, A. (2021). Young consumers’ intention to par- 189. https://doi.org/10.1080/0965254x.2017.1291173. ticipate in The sharing economy: An integrated mod- Jelinkova, M., Tetrevova, L., Vavra, J., & Munzarova, S. el. Sustainability, 13(1), 430. https://doi.org/10.3390/ (2021). The sharing economy in the context of Sus- su13010430. tainable Development and Social Responsibility: The Mazurek, S. (2020). The Professionalization of Crowd- example of the Czech Republic. Sustainability, 13(17), sourcing: A Process Transforming Shared Economy. 9886. https://doi.org/10.3390/su13179886. Transformations in Business & Economics, 19, 429– Johnson, T. P. (2014). Snowball sampling: Introduction. 447. Wiley StatsRef: Statistics Reference Online. https://doi. Ministry of the Interior of the Czech Republic. Informa- org/10.1002/9781118445112.stat05720. tivní počty obyvatel v obcích - Ministry of the Interior Khare, A., Khare, A., & Singh, S. (2012). Attracting shop- of the Czech Republic. Retrieved August 2, 2023, from pers to shop online—challenges and opportunities for https://www.mvcr.cz/clanek/informativni-pocty-oby- the Indian retail sector. Journal of Internet Commerce, vatel-v-obcich.aspx. 11(2), 161–185. https://doi.org/10.1080/15332861.201 Moeller, S., & Wittkowski, K. (2010). The burdens of 2.689570. ownership: Reasons for preferring renting. Managing Kirchherr, J., & Charles, K. (2018). Enhancing the sam- Service Quality: An International Journal, 20(2), 176– ple diversity of Snowball Samples: Recommenda- 191. https://doi.org/10.1108/09604521011027598. tions from a research project on anti-dam movements Molek, N., Markič, M., Janežič, D., Lužar, M., & Brcar, in Southeast Asia. PLOS ONE, 13(8). https://doi. F. (2023). Leadership Styles and Generational Differ- org/10.1371/journal.pone.0201710. ences in Manufacturing and Service Organizations. Kotler, P., & Armstrong, G. M. (2010). 2. Pearson Educa- Organizacija,56(3) 221-232. https://doi.org/10.2478/ tion India. orga-2023-0015. Lakens, D. (2013). Calculating and reporting effect sizes Nadler, J. T., Weston, R., & Voyles, E. C. (2015). Stuck in to facilitate cumulative science: A practical primer the Middle: The Use and Interpretation of Mid‐Points for T-tests and ANOVAS. Frontiers in Psychology, 4. in Items on Questionnaires. The Journal of General https://doi.org/10.3389/fpsyg.2013.00863. Psychology, 142(2), 71–89. https://doi.org/10.1080/00 Lazakidou, A., Ilioudi, S., & Siassiakos, K. (2008). Elec- 221309.2014.994590. tronic customer relationship management applications Parameswaran, M., & Whinston, A. B. (2007). Social and consumer trust in e-commerce. International Jour- Computing: An Overview. Communications of the As- nal of Electronic Customer Relationship Management, sociation for Information Systems, 19(37) https://doi. 2(3), 262. https://doi.org/10.1504/ijecrm.2008.020411. org/10.17705/1CAIS.01937 Lee, S., Lee, W. J., & Yoo, K.-H. (2019). Millennial ride- Pham, H.T., Hoang, K.T., Nguyen, T.T., Do, P., & Mar, share passengers’ pro-sustainable behaviours: Norm M.T. (2021). Sharing Economy: Generation Z’s Inten- Activation Perspective. Asia Pacific Journal of Tour- tion Toward Online Fashion Rental in Vietnam. Jour- ism Research, 25(1), 15–26. https://doi.org/10.1080/1 nal of Asian Finance, Economics and Business, 8, 997- 0941665.2019.1578811. 1007. https://doi.org/10.13106/JAFEB.2021.VOL8. López-Martín, E., & Ardura-Martínez, D. (2023) The ef- NO3.0997. fect size in scientific publication. Educación XX1, Prothero, A., Dobscha, S., Freund, J., Kilbourne, W. E., 26(1), 9-17. https://doi.org/10.5944/educxx1.36276. Luchs, M. G., Ozanne, L. K., & Thøgersen, J. (2011). Lucian, R. (2016). Rethinking the use of Likert scale: tra- Sustainable consumption: Opportunities for consumer dition or technical choice? Brazilian Journal of Mar- research and public policy. Journal of Public Policy keting, Opinion, and Media Research, 9(1), 11–26. & Marketing, 30(1), 31–38. https://doi.org/10.1509/ Mahadevan, R. (2018). Examination of motivations and jppm.30.1.31. attitudes of peer-to-peer users in the accommodation Ranzini, G., Newlands, G., Anselmi, G., Andreotti, A., sharing economy. Journal of Hospitality Marketing & Eichhorn, T., Etter, M., Hoffmann, C. P., JJrss, S., & Management, 27(6), 679–692. https://doi.org/10.1080/ Lutz, C. (2017). Millennials and the sharing econo- 19368623.2018.1431994. my: European perspectives. SSRN Electronic Journal. Mangiafico, S. S. (2016). Summary and Analysis of Ex- https://doi.org/10.2139/ssrn.3061704. tension Program Evaluation in R, version 1.20.01, re- Richardson, J. T. E. (2011). Eta squared and partial eta trieved from https://rcompanion.org/, last access June squared as measures of effect size in educational re- 21, 2023. search. Educational Research Review, 6(2), 135-147. 331 Organizacija, Volume 57 Research Papers Issue 4, November 2024 https://doi.org/10.1016/j.edurev.2010.12.001. doi.org/10.1108/ijchm-09-2016-0496. Rossmannek, O., & Chen, M. (2023). Why people use The sharing economy: A meta-analysis. Journal of Clean- er Production, 387, 135824. https://doi.org/10.1016/j. Pavel Pelech is doctoral student at the Technical jclepro.2022.135824. University of Liberec in the Department of Marketing. He Sahelices‐Pinto, C., Lanero‐Carrizo, A., & Vázquez‐Bur- has more than 20 years of experience in development guete, J. L. (2020). Self‐determination, clean con- management, marketing and company management. science, or social pressure? underlying motivations His expertise is not just limited to practical experience for organic food consumption among young millenni- but also extends to academic contributions. He has als. Journal of Consumer Behaviour, 20(2), 449–459. published four research papers that provide new https://doi.org/10.1002/cb.1875. insights into marketing communications and P2P Schewe, C. D., & Meredith, G. (2004). Segmenting glob- business. These papers look at the impact of the al markets by generational cohorts: determining mo- sharing economy on marketing communications from several perspectives. His papers set the stage for tivations by age. Journal of Consumer Behaviour, 4, future research in these areas. 51–63. https://doi.org/10.1002/cb.157. Sturgis, P., Roberts, C., & Smith, P. (2012). Middle alterna- tives revisited. Sociological Methods & Research, 43(1), Jaroslava Dědková is an associate professor at the 15–38. https://doi.org/10.1177/0049124112452527. Faculty of Economics of the Technical University of Tu, Y.-T., Aljumah, A. I., Van Nguyen, S., Cheng, C.-F., Liberec, where she is the head of the Department Tai, T. D., & Qiu, R. (2023). Achieving Sustainable of Marketing and Business. Her teaching activity is Development Goals through a sharing economy: Em- focused on Strategic Marketing, Consumer Behaviour, pirical evidence from developing economies. Journal and Regional Marketing in Tourism. She is a member of Innovation & Knowledge, 8(1), 100299. https://doi. of the Section of Teachers of Marketing at the Czech org/10.1016/j.jik.2022.100299. Marketing Society. She supervises doctoral students Yang, S., & Ahn, S. (2016). Impact of motivation in the and supervises master theses and bachelor theses. sharing economy and perceived security in attitude and She is a member of the examination committees at the loyalty toward Airbnb. Advanced Science and Technol- Faculty of Economics, TU Liberec. In her scientific and ogy Letters. https://doi.org/10.14257/astl.2016.129.36 research activities she focuses on innovative marketing, Yusoff, M. S., Arifin W. N., & Hadie S. N. H. (2021). ABC consumer behaviour and regional marketing. She is of questionnaire development and validation for sur- a researcher or co-investigator of several projects vey research. Education in Medicine Journal, 13(1), with cross-border impact. She publishes in scientific 97–108. https://doi.org/10.21315/eimj2021.13.1.10. journals and conference proceedings. She regularly Zhu, G., So, K. K., & Hudson, S. (2017). Inside the shar- completes teaching internships at foreign partner universities within the Erasmus+ project. ing economy. International Journal of Contemporary Hospitality Management, 29(9), 2218–2239. https:// Preučevanje generacijskih razlik v ekonomiji delitve: razumevanje motivacij lastnikov nedejavnih sredstev in njihov vpliv na upravljanje enakovrednih platform Ozadje/Namen: Raziskava preučuje dinamiko deljenja nedejavnih sredstev na enakovrednih platformah (P2P) znot- raj ekonomije delitve s poudarkom na razumevanju motivacijskih dejavnikov, ki vplivajo na lastnike sredstev med ge- neracijami. Platforme, ki razumejo, kaj motivira lastnike sredstev, da vstopijo v ekonomijo delitve, lahko učinkoviteje vzpostavijo in upravljajo tržno komuniciranje ter pridobijo konkurenčno prednost. Metode: Uporabljen je bil pristop mešanih metod, ki vključuje kvantitativno in kvalitativno zbiranje ter analizo podat- kov. Analitična orodja vključujejo opisno statistiko, Kruskal-Wallisov test, PCA in Welchovo ANOVA. Rezultati: Identificirani so bili štirje motivacijski dejavniki. Ugotovljeno je bilo, da je generacijsko trženje primerno za tržna sporočila, ki ciljajo na oportunistične dejavnike (popolnoma nove) ali družbene dejavnike. Za uporabniške in družbene dejavnike generacijsko trženje ni potrebno, tj. platformam P2P ni treba razlikovati segmenta, na katerega ciljajo s tržno komunikacijo. Zaključek: Raziskava razkriva, kako generacijske razlike vplivajo na motivacijo za sodelovanje v ekonomiji delitve, kar pomaga pri upravljanju platform P2P. Ponuja novo primerjalno analizo generacijskih motivatorjev, bogati litera- turo o ekonomiji delitve in zagotavlja praktične vpoglede za učinkovito ciljanje na različne generacije. Razumevanje motivacije lastnikov nedejavnih sredstev je ključno za uspešno upravljanje poslovanja platform P2P. Ključne besede: Vedenje potrošnikov, Generacije, Management, Trženje, P2P, Delitvena ekonomija 332 Organizacija, Volume 57 Research Papers Issue 4, November 2024 DOI: 10.2478/orga-2024-0024 The Use of Artificial Intelligence among Students in Higher Education Eva JEREB, Marko URH University of Maribor, Faculty of Organizational Sciences, Kranj, Slovenia, eva.jereb@um.si, marko.urh@um.si Background/Purpose: Artificial intelligence (AI) impacts our everyday lives, from business to social areas, and, in recent years, more and more education. This paper aims to discuss using artificial intelligence tools for educational purposes from the student’s perspective. Methods: A quantitative approach was used for the research using the questioning method. Data were collected using an e-survey. The questionnaire contained closed questions referring to (i) general data, (ii) the use of specific artificial intelligence tools, and open questions (iii) about using artificial intelligence for study purposes. Parametric Independent – Samples t-Test and One-Way ANOVA tests were used for normal and near-normal distributions of the responses. Results: The study conducted with 258 university students in Slovenia reveals a general familiarity with AI concepts yet with a limited ability to articulate this understanding. Popular AI tools like ChatGPT and Grammarly are predomi- nantly used for information retrieval and idea generation. Notable differences in AI tool usage were observed based on gender and field of study. Conclusion: The rapid advancement of AI is significantly transforming higher education. Integrating AI into educa- tion fosters the development of adaptive, personalized, and inclusive learning environments. Due to the study’s lim- ited sample size and geographic focus, further research with more diverse samples is needed to understand global AI tool usage in higher education fully. Keywords: Artificial intelligence, AI tools, Higher education, Students, Skills 1 Introduction nity is already trying to find ways to successfully imple- ment AI for staff and students (Barrett et al., 2019). Many Artificial intelligence (AI) is influencing all areas of authors like Ahmand et al. (2021), Crompton and Burke our lives today. It is used across various industries and ap- (2023), Dahri et al. (2023), Chen et al. (2022), Sultana and plications, transforming how tasks are performed. It helps Faruk (2024), and others have dealt with artificial intelli- to solve complex problems in healthcare, finance, retail, gence in the field of education. Using AI in education goes transportation, manufacturing, customer service, market- beyond adopting technologies to facilitate easier learning. ing, cybersecurity, human resources, legal, and many other It also means reshaping, redesigning, and rethinking tra- areas. Artificial intelligence technology is becoming the ditional education systems’ content and methods. AI can basis for business (Barrett et al., 2019). It is developing be applied in education through three fundamental models quickly and has enormously impacted today’s world. AI (Luckin & Holmes, 2016): pedagogical model (knowledge technologies are beneficial not only to the business sector and expertise of teaching), domain model (knowledge of but also to the educational domain. The education commu- the subject being learned - domain expertise), and learner 1 Received: 21st June 2024; Accepted: 9th September 2024 333 Organizacija, Volume 57 Research Papers Issue 4, November 2024 model (knowledge of the learner). These models were de- and planning. Specifically, current definitions suggest that veloped to represent emotional, social, and metacognitive intelligence is the ability to learn from experience and rec- aspects of learning in education (Shen et al., 2021). ognize and solve problems. The acquisition, retention, and Many benefits of using AI in education can improve use of knowledge is a vital intelligence component. To use the student’s learning (Singh & Mishra, 2023). For ex- knowledge, one must first identify the issues it might ad- ample, a significant advantage is personalized education. dress and then use what he has learned to solve problems The current educational system relies on a one-size-fits- (Jaarsveld & Lachmann, 2017). all model, but no two people have the same skills. AI can People are termed as born intelligent. However, ma- provide a more individualized learning experience and chines are created to resolve what is outlined in their pro- create lesson plans to teach each student effectively. It can gramming. Once a machine can show intelligence like any also comprehend how a student is learning. AI can iden- other human being, it is called artificial intelligence (Aro- tify places where students are struggling and offer them ra, 2021). Artificial intelligence is the simulation of human alternate learning strategies. The AI algorithm analyzes intelligence processes by machines, especially computer students’ learning behavior and adapts the course contents systems (Laskowski & Tucci, 2023). We can say that ar- to support each student. Then, we have task automation, tificial intelligence refers to the development of computer monitoring, and feedback. AI can help identify grammati- systems that can perform tasks that typically require hu- cal errors, sentence structure problems, and more. man intelligence. These tasks include learning, reasoning, AI methods have also revolutionized teachers’ work problem-solving, perception, language understanding, and by helping them track student performance using various speech recognition. AI aims to create machines or software tools, including automated grading, learning analytics, and that mimic human cognitive functions and sometimes sur- adaptive learning platforms. Automated grading systems pass human capabilities. offer quick, reliable assessments, saving teachers time AI can be categorized into narrow or weak AI and gen- and providing timely feedback to students. Learning an- eral or strong AI (AGI). Narrow AI is designed and trained alytics reveal student behavior and performance patterns, for a particular task. It excels in performing specific func- enabling educators to adjust their teaching methods based tions but lacks the broad cognitive abilities of a human. on these insights. Predictive analytics analyzes historical Weak AI is often focused on performing a single task ex- data to foresee potential challenges, allowing educators to tremely well. While these machines may seem intelligent, intervene and support students proactively (Chandrasekar, they operate under far more constraints and limitations 2023; Kim, 2024; Rensfeldt & Rahm, 2023) Monitoring than basic human intelligence (Schroer, 2023). Examples and feedback will help teachers understand how their stu- include virtual personal assistants, image and speech rec- dents receive education so that they can help them and also ognition systems, and recommendation algorithms. Artifi- see the gaps in the curriculum that need to be filled or im- cial general intelligence, often portrayed in science fiction, proved (Seraydarian, 2021). refers to machines that can understand, learn, and apply Because artificial intelligence is progressing at an knowledge across a wide range of tasks at a human level. accelerated pace in the education domain, we wanted to Like human general intelligence, AGI would have many research how familiar students are with artificial intelli- advantages compared to narrow (limited, weak, special- gence. Are they familiar with the AI concept? Which AI ized) AI. An AGI system would be much more flexible and platforms and tools do they use, and how often? How adaptive. AGI systems also require fewer human interven- skilled are they in using them? Do they think using arti- tions to accommodate the various loose ends among partial ficial intelligence helps them better understand the study elements, facets, and perspectives in complex situations content? Where does artificial intelligence benefit them (Korteling et al., 2021). most in their study? Do they believe that educational insti- Both narrow AI and AGI concepts have been explored tutions should offer training in using artificial intelligence and applied in education, but most practical implementa- for studying? We also wanted to know if there are differ- tions fall under the narrow AI category. For example, Per- ences between students using AI tools regarding gender sonalized Learning Platforms (PLP) are AI systems that and the study field. analyze students’ learning patterns and provide custom- ized study materials, exercises, and pacing recommenda- 2 Theoretical Background tions. Many authors like Hashim et al. (2023), Farooq et al. (2024), Naseer et al. (2023), and others are exploring personalized learning platforms. All the students have dif- The ability to accumulate knowledge and then use it ferent aptitudes, learning skills, and orientations. With AI’s for solving problems is generally termed intelligence. Al- assistance, only content required and suited to the student though contemporary definitions of intelligence vary con- is delivered (Arora, 2021). Then, we have Intelligent Tu- siderably, experts generally agree that intelligence involves toring Systems (ITS), which use AI to adapt the learning mental abilities such as logic, reasoning, problem-solving, experience based on students’ individual needs. They can 334 Organizacija, Volume 57 Research Papers Issue 4, November 2024 provide feedback, answer questions, and guide students simulates a human conversation with an end user. Though through learning. ITS replicates teachers’ roles and in- not all chatbots are equipped with artificial intelligence, creasingly automates pedagogical functions (e.g., problem modern chatbots increasingly use conversational AI tech- generation, problem selection, and feedback generation) to niques like natural language processing (NLP) to under- help create new methods and redefine educational goals stand the user’s questions and automate responses (IBM, (Shen et al., 2021). Lin (2023), Ramadhan (2024), Rybi- 2023). ChatGPT is a very well-known chatbot today, de- na (2023), and many others are writing about intelligent veloped by OpenAI. Many authors like Garrel and May- tutoring systems. Language Processing Applications: Nat- er (2023), Bhullar et al. (2024), Korseberg and Elken ural Language Processing (NLP) (Campino, 2024; Kao- (2024), Jensen et al. (2024), Rawas (2024) and others are uni, 2024; Wu, 2024) is used in educational tools for tasks researching the use of ChatGPT in education. Based on a like automated grading of essays, language learning apps, large language model, it enables users to refine and steer a and chatbots that assist students with queries. In the edu- conversation towards a desired length, format, style, level cation sector, chatbots are also used in different processes of detail, and language. Similar to it is Bing, a web search like admission, career counseling, etc. (Malik & Solanki, engine owned and operated by Microsoft. Bard is a con- 2021). In Learning Analytics, AI is applied to analyze data versational generative artificial intelligence chatbot devel- generated by students’ interactions with digital learning oped by Google, based initially on the LaMDA (Language platforms. This information helps educators understand Model for Dialogue Applications), PaLM, (Pathways student progress and identify areas needing additional at- Language Model), and Gemini families of large language tention. models. Many others exist, such as Jasper Chat, Claude 2, Many systems, platforms, or tools are available to stu- Llama 2, HuggingChat, etc. Also well-known is Perplexit- dents, helping them in their educational process. Thinkers yAI, a user interface strategy similar to ChatGPT, but it is on mathematics were designed to help students with any less a chatbot and more of a search bot. It closely resem- help in math. They relate maths with real-life situations bles the Google search engine in its layout and includes a and are also appropriate for kids of a young age. For exam- prominent central search bar where users can input their ple, Math Thinker was designed for free for K-12 students questions to the AI. to have fun and help in math, but it is also suitable for older Brainly is the knowledge-sharing community where students. It was initiated because of the problems during hundreds of millions of students and experts put their the Covid-19 school shutdowns (https://maththinker.org/ heads together to crack their most challenging tasks about.html). Thinkster Math provides the student with spe- (https://brainly.com/). Mika is a personalized virtual tutor cific issues fitting their scope of capacities and abilities. that easily adapts to student’s needs and provides real-time It upgrades explanations by providing video help (https:// feedback to help them learn more effectively. SmartEd al- hellothinkster.com/). Authors like Casal-Otero et al. lows students to easily customize learning materials such (2023), Lane (2023), Wang and Lester (2023), and many as textbooks to the student’s learning style and needs. It others are researching the use of AI in K-12 education. also has gamification features that make learning more en- Some platforms support students’ work on a specific gaging and fun (Seraydarian, 2021). text. They help to learn and master content by breaking text down into manageable pieces of information. For ex- ample, Cram101’s AI technology can turn any textbook 3 Method into an intelligent study guide complete with chapter sum- maries, unlimited true-false and multiple-choice practice 3.1 Sample tests, and flashcards drilled down to a specific book, ISBN, author, and chapter (https://contenttechnologiesinc.com/). The study sample consisted of 258 students from ran- JustTheFacts101 highlights and generates text and chap- domly selected faculties of universities in Slovenia. Four- ter-specific summaries on the spot. Platforms that check teen did not answer the general questions about gender the grammatical correctness of the text, as well as the ap- and study level. Of the 244, 46.3 % were male and 53.7 propriate formation of whole sentences, are widely used % were female; 65.6 % were undergraduate and 34.4 % today. For instance, Grammarly instantly generates clear postgraduate students. Sixteen did not answer the general and compelling writing while maintaining the student’s questions about the study field. Of the 242 students, 55.4% unique voice (https://www.grammarly.com/). Platforms were social science students, 12.4 % were natural science like ProWritingAid, Hemingway App, WhiteSmoke, and students, and 32.2 % were technical sciences students (for WordTune, all similar to Grammarly, help students im- more details, see Table 1). prove spelling, grammar, style, and impact. Very often used today are chatbots. A chatbot is a software application or web interface designed to mimic human conversation through text or voice interactions. It 335 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 1: Frequency distributions of the study variables (n=258) Frequency Percent Valid percent Male 113 43.8 46.3 Gender Female 131 50.8 53.7 Missing 14 5.4 Bachelor 160 62 65.6 Study level Masters 84 32.6 34.4 Missing 14 5.4 Social sciences 134 51.9 55.4 Study field Natural sciences 30 11.6 12.4 Technical sciences 78 30.2 32.2 Missing 16 6.2 3.2 Questionnaire and Procedure heard of the term artificial intelligence, were familiar with the basic concepts of AI, and had ever encountered tools The questionnaire contained closed questions refer- using AI. 255 (99.2 %) students had already heard of the ring to (i) general data (gender, level of study, and field of term AI, and 2 (0.8 %) had not. 229 (88.8 %) answered study), (ii) the use of specific artificial intelligence tools, that they know the basic concepts of AI, and 29 (11.2 %) and open questions (iii) about using artificial intelligence that they do not. 229 (88.8 %) had encountered using tools for study purposes. or apps that involve artificial intelligence, thirteen (5 %) The frequency of using specific AI tools was meas- did not, and 16 (6.2 %) did not know if they had (Table 2). ured on a 5-point Likert-type scale. Students were asked to We also wanted to know if there are differences be- choose from “1=Never”, “2=Rarely”, “3=Medium often”, tween students using AI tools regarding gender and the “4=Often”, and “5=Very often”. The answers regarding study field. First, we tested if there are any differences be- the helpfulness of using the AI tools in different areas of tween males and females regarding knowing the concepts their studies were measured on a 5-point Likert-type scale. of artificial intelligence. No statistically significant differ- Students were asked to choose from “1=Not helpful”, ences were found (t = -1.459 and p = 0.146). There were “2=A little helpful”, “3=Partially helpful”, “4=Helpful”, also no statistically significant differences between stu- and “5=Very helpful”. dents of different study fields regarding knowing the con- The skills in using AI tools were measured on a 5-point cepts of artificial intelligence (F = 2.429 and p = 0.090). Likert-type scale from “1= Not very skilled”, “2=A little The second set of questions concerns the use of skilled”, “3=Partially skilled”, “4=Skilled”, and “5=Very artificial intelligence tools. First, we asked students if they skilled”. Students were asked to rate their level of satisfac- use apps or tools, including artificial intelligence, for study tion with the use of AI tools on a 5-point Likert-type scale (learning) purposes. Within this question, we also asked from “1= Dissatisfied”, “2=Not too pleased”, “3=Satis- them to skip the next set of questions if the answer was no. fied”, “4=Very satisfied”, and “5= Excited”. Two hundred fifty students answered this question, where The online questionnaire was presented to students by 187 (75 %) use AI tools for study purposes, and 63 (25 %) professors during lectures and tutorials. Participation in do not. Of 185 students who answered the next question, this research was voluntary and anonymous. 43 (23 %) use AI tools rarely, 81 (44 %) use AI several All statistical tests were performed with SPSS 27. Par- times a month, 44 (24 %) several times a week, and 17 (9 ametric Independent – Samples t-Test and One-Way ANO- %) use AI tools almost every day (Figure 1). Two students VA tests were used for normal and near-normal distribu- didn’t answer. We tested if there were any differences be- tions of the responses. tween males and females regarding the frequency of using AI tools. No statistically significant differences were found (t = 1.796 and p = 0.074). There were also no statistically 4 Results significant differences between students of different study fields regarding the frequency of using AI tools (F = 2.520 The first set of questions concerns general knowledge and p = 0.083). about artificial intelligence. We asked students if they had 336 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 2: Descriptive statistics for the first set of questions Yes No Do not know Question N Freq. Percent Freq. Percent Freq. Percent Have you heard of the term artifi- cial intelligence? 257 255 99.2 2 0.8 / / Are you familiar with the basic concepts of artificial intelligence? 258 229 88.8 29 11.2 / / Have you ever encountered using apps or tools that involve AI? 258 229 88.8 13 5.0 16 6.2 Figure 1: The use of AI tools for study purposes in percentage Next, we asked them which AI tools they use and how payable AI tools. We found statistically significant differ- often. The results can be seen in Table 3. Of all the tools, ences between genders regarding using both payable and students use ChatGPT and Grammarly the most. Some stu- free AI tools (t = 3.183 and p = 0.002). Males (M = 0.2093 dents also use other AI tools, but rarely. These tools are and SD = 0.4092) are more likely to use payable and free MS Bing, Perplexity AI, Google Bard, Jasper Chat, Chat- AI tools than females (M = 0.015 and SD = 0.2223). We Sonic, Claude 2, Llama 2, HuggingChat, ProWtitingAid, also found statistically significant differences between Hemingway App, WordTune, and Midjourney. genders regarding using only free AI tools (t = -2.919+ and Since ChatGPT and Grammarly were the tools used p = 0.004). Females (M = 0.9381 and SD = 0.2421) are most often among students, we tested if there were any likelier to use only free AI tools than males (M = 0.7907 significant differences between males and females and stu- and SD = 0.4092). We found statistically significant differ- dents of different study fields. No statistically significant ences between technical and social students using payable differences were found between genders (ChatGPT: T = and free AI tools. The Games-Howell test showed that stu- -0.350 and p = 0.727; Grammarly: t = 0.460 and p = 0.646) dents from the technical field are more likely to use pay- and between students of various study fields (ChatGPT: able and free AI tools than students from the social field F = 0.180 and p = 0.836; Grammarly: F = 1.134 and p = (Sig = 0.016). Students from the social field are more like- 0.324). ly to use only free AI tools than technical students (Sig = The next question addressed using payable or free AI 0.031). No statistically significant differences were found tools. One hundred eighty-six students answered the ques- between students from the social and natural fields and stu- tion, 161 (86.6 %) use only free AI tools, 24 (12.9 %) use dents from the technical and natural fields regarding using both free and payable AI tools, and one student uses only payable or free AI tools. 337 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 3: Descriptive statistics for the use of AI tools (Mean and Std. Deviation) Never Rarely Medium often Often Very often N Mean Std. de- viation ChatGPT 6 (3%) 34 (18%) 51 (28%) 54 (29%) 40 (22%) 185 3.5 1.1 MS Bing 149 (84%) 11 (6%) 10 (6%) 3 (2%) 4 (2%) 177 1.3 0.8 Perplexity AI 155 (88%) 13 (7%) 6 (3%) 2 (1%) 1 (1%) 177 1.2 0.6 Google Bard 148 (84%) 21 (12%) 5 (3%) 2 (1%) 0 (0%) 176 1.2 0.5 Jasper Chat 166 (94%) 7 (4%) 2 (1%) 0 (0%) 1 (1%) 176 1.1 0.4 ChatSonic 166 (95%) 5 (3%) 3 (2%) 0 (0%) 1 (1%) 175 1.1 0.4 Claude 2 164 (95%) 5 (3%) 2 (1%) 1 (1%) 1 (1%) 173 1.1 0.5 Llama 2 171 (98%) 3 (2%) 1 (1%) 0 (0%) 0 (0%) 175 1 0.2 HuggingChat 170 (98%) 3 (2%) 0 (0%) 0 (0%) 0 (0%) 173 1 0.1 Grammarly 86 (49%) 34 (19%) 23 (13%) 26 (15%) 6 (3%) 175 2 1.2 ProWritingAid 165 (96%) 4 (2%) 2 (1%) 0 (0%) 0 (0%) 171 1 0.3 Hemingway App 168 (98%) 3 (2%) 1 (1%) 0 (0%) 0 (0%) 172 1 0.2 WhiteSmoke 169 (98%) 2 (1%) 1 (1%) 0 (0%) 0 (0%) 172 1 0.2 WordTune 161 (95%) 6 (4%) 2 (1%) 0 (0%) 0 (0%) 169 1.1 0.3 Midjourney 157 (92%) 7 (4%) 3 (2%) 2 (1%) 2 (1%) 171 1.2 0.6 1=Never, 2=Rarely, 3=Medium often, 4=Often, and 5=Very often Table 4: Descriptive statistics for skills in using AI tools How skilled are you in using AI tools? Frequency Percent Valid Percent Cumulative Percent Valid Not very skilled 25 9,7 13,4 13,4 A little skilled 50 19,4 26,9 40,3 Partially skilled 71 27,5 38,2 78,5 Skilled 34 13,2 18,3 96,8 Very skilled 6 2,3 3,2 100,0 Total 186 72,1 100,0 Missing Prekinjeno 8 3,1 Leap (if) 64 24,8 Total 72 27,9 Total 258 100,0 338 Organizacija, Volume 57 Research Papers Issue 4, November 2024 We also asked students how skilled they are using AI the areas where using AI tools benefits them. However, tools. We can see that most students (56.5 %) are partially we found statistically significant differences between un- skilled or skilled in using AI tools. More details can be dergraduate and postgraduate students regarding faster seen in Table 4. The mean value for the above question is learning (t = 2.218 and p = 0.014) and task solving (t = 2.71, with a standard deviation of 1.020. We found statis- 1.998 and p = 0.047). Undergraduate students find AI tically significant differences regarding skills of using AI tools more helpful for faster learning (M = 3.4 and SD = tools between genders (t = 3.881 and p < 0.001) and also 1.076) than postgraduate students (M = 2.95 and SD = between students of various study fields (F = 6.605 and 1.371). Undergraduate students also find AI tools more p = 0.002). Males (M = 3.01 and SD = 1.035) are more helpful for task solving (M = 3.01 and SD = 1.185) than skilled in using AI tools than females (M = 2.44 and SD = postgraduate students (M = 2.63 and SD = 1.139). 0.946). Most skilled in using AI tools are students from the We also asked them if they think using artificial intel- technical field (M = 3.02 and SD = 0.976), then students ligence helps them better understand the study content. from the natural field (M = 2.88 and SD = 0.993), and the Of 183 students, 158 (86.3 %) think AI helps them better least skilled are students from the social field (M = 2.44 understand the study content, and 25 (13.7 %) do not. and SD = 1.018). The third and last set of questions concerns the issue The next question addressed satisfaction with using AI of involving artificial intelligence tools in the educational tools for study purposes. Of 186 students, 20 (10.8 %) are process. The questions and answers can be seen in Table excited about using AI tools, 33 (17.7 %) are very satisfied, 6. 120 (64.5 %) are satisfied, 12 (6.5 %) are not too pleased, and one is dissatisfied. We can see that the majority of stu- dents are satisfied or even more. The mean value regarding 5 Discussion satisfaction with using AI is 3.32, with a standard devia- tion of 0.772. We found no statistically significant differ- Most students reported familiarity with AI concepts, ences regarding satisfaction with using AI tools for study which aligns with AI’s growing awareness and integration purposes between genders (t = 1.562 and p = 1.120) and into various aspects of society. However, their ability to between students of various study fields (F = 0.203 and p articulate this understanding in their own words was lim- = 0.817). ited. The best explanation, in the words of a student, was Next, we wanted to know where the AI benefits them “a software tool that “takes” data from a wide database, most in their study. The results can be seen in Table 5. We according to our “requirements.””. Students’ statements can see that the areas where AI tools help students the most about missing “a single AI that can use text and images are searching for information (M = 3.8 and SD = 1.1) and and can take a picture of text and have it automatically generating ideas (M = 3.6 and SD = 1.1). written” or “AI tools for simplifying very general and broad Statistically significant differences between genders topics, so there’s no need to search for information and were revealed only for generating ideas (t = -2.325 and read entire books or articles” indicate that some are unfa- p = 0.021). There were no statistically significant differ- miliar with AI tools and their capabilities. ences between students of various study fields regarding Table 5: Descriptive statistics of areas where using AI tools benefits students Not helpful A little hel- Partially pful helpful Helpful Very helpful N Mean Std. devia- tion Faster Learning 18 (10%) 31 (17%) 55 (30%) 48 (26%) 31 (17%) 183 3.2 1.2 Generating ideas 6 (3%) 20 (11%) 51 (28%) 62 (34%) 42 (23%) 181 3.6 1.1 Writing seminar papers 17 (9%) 45 (25%) 63 (35%) 31 (17%) 25 (14%) 181 3 1.2 Searching for information 8 (4%) 9 (5%) 48 (26%) 65 (35%) 54 (29%) 184 3.8 1.1 Explanation of material 16 (9%) 33 (18%) 53 (29%) 46 (25%) 35 (19%) 183 3.3 1.2 Tasks solving 25 (14%) 43 (24%) 62 (34%) 28 (15%) 23 (13%) 181 2.9 1.2 Problem-solving 15 (8%) 39 (21%) 65 (36%) 38 (21%) 25 (14%) 182 3.1 1.1 339 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 6: Descriptive statistics for the third set of questions Yes No Do not know Question N Freq. Percent Freq. Percent Freq. Percent Do you think using artificial intelli- gence positively affects the quality 185 156 84.3 29 15.7 / / of your study? Do you think that artificial intelli- gence will be even more involved in 184 169 91.8 3 1.6 12 6.5 educational processes in the future? Do you think students should learn more about how artificial intelli- 184 155 84.2 11 6.0 18 9.8 gence works during their studies? Do you think that educational insti- tutions should offer training in the field of using artificial intelligence 183 146 79.8 16 8.7 21 11.5 for studying? ChatGPT and Grammarly emerged as the most fre- Students’ moderate level of skills in using AI tools is quently used AI tools, which resonates with broader trends. consistent with the learning curve associated with adopt- Natural language processing tools and writing assistance ing new technologies (Zehng et al., 2021; Sumakul et al., applications are widely adopted due to their versatility and 2022; Coffey, 2023; Salido, 2023). Further studies might direct relevance to academic tasks. A few students listed to delve into the specific skills students find challenging or the majority lesser-known AI tools like Kapwing, a known proficient in using AI tools. online video editing platform. It has gained popularity, es- The predominant use of AI tools among students was pecially among users who seek easy-to-use tools for creat- for information retrieval and idea generation. The per- ing and editing multimedia content. Next was DALL-E, an ceived benefits of AI tools in faster learning, idea gen- AI model OpenAI developed that generates images from eration, and information retrieval are also discussed by textual descriptions. It gained significant attention due to authors like Delcker et al. (2024), Bello (2024) and oth- its ability to create unique and imaginative images based ers. Students also use AI tools for writing seminar papers, on user prompts. However, it may not be as widely used task-solving, problem-solving, material simplification, and by the general public as some more consumer-oriented ap- a more straightforward understanding of matter and tasks. plications. Students also mentioned they use the Orange One student wrote he uses AI tools for reading and gener- Data Mining AI tool, an open-source data visualization ating poetry. and analysis tool. It is known in the data science and ma- Most students express satisfaction and excitement with chine learning communities for its user-friendly interface AI tool usage, which aligns with the generally positive atti- and versatility. It’s famous for educational purposes and tude towards technology adoption in education. However, exploratory data analysis. Some students also use rewrit- specific factors contributing to satisfaction may vary and ing AI tools or services with text-rewriting capabilities for could be explored further. academic purposes. They did not reveal which ones. The anticipation of increased AI involvement in edu- The study identified differences in using payable AI cational processes aligns with the overall trajectory of AI tools and skills based on gender. Males are likelier to use adoption in education globally. Education with the help of payable and free AI tools than females. While this aligns artificial intelligence is most developed in several coun- with the general acknowledgment of gender gaps in tech- tries, each focusing on different aspects of AI integration nology adoption, the extent of these differences may vary in education. Notable countries leading in AI education across different studies and cultural contexts. Students are the United States, China, Singapore, South Korea, from the technical field are more likely to use payable and Finland, Spain, Japan, Sweden, and Luxembourg (Bak- free AI tools than students from the social field. Technical er, 2017; Lake, 2023; Basheer, 2024; Hanks, 2024). In science students showing a higher propensity to use also terms of specific fields, AI is being used broadly across payable AI tools and demonstrating greater proficiency various educational domains, including (STEM) science, align with the expectation that specific disciplines might technology, engineering, mathematics, robotics, language engage more deeply with AI technologies. learning, special education, and administrative efficiency 340 Organizacija, Volume 57 Research Papers Issue 4, November 2024 (World Economic Forum, 2024; Baker, 2017; Lake, 2023; large datasets efficiently suggests it could handle tasks tra- Hanks, 2024). ditionally performed by human researchers. Studies often reflect an awareness of the evolving role They know that AI contributes only a part in making it of AI in shaping educational practices. Most students sup- easier and faster for humans to perform tasks. One student porting the incorporation of AI training into educational stated, “I firmly believe that we must make an effort for curricula are consistent with recognizing AI as a crucial everything ourselves, and artificial intelligence can only skill set for future professionals. This aligns with the serve as assistance, as Google did initially, significantly broader discourse on preparing students for the AI-driven easing everything for us.” One other said, “It would be workforce. right for people to be acquainted with artificial intelligence Students stated that since there are many valuable ar- and use it for assistance, not exploitation.” Students are tificial intelligence tools, they should learn to use as many aware that some students are exploiting AI tools with min- AI tools as possible at the faculty. They want to involve imal effort to perform different tasks regarding study. artificial intelligence more in their studies and use it to as- Based on the results and students’ statements, we sug- sist them in various projects they must complete at facul- gest that educational institutions should offer training in ties. They want to learn how to use artificial intelligence using AI for studying purposes. most skillfully. They think it would improve the quality of learning and reduce the time needed for specific tasks related to the study purposes. One student stated, “Instead 6 Conclusion of suppressing the use of AI, we should use it to our advan- tage, such as obtaining key data.” Another statement was, Artificial intelligence is progressing at an accelerated “It would be excellent if we could use artificial intelligence pace, which already impacts the profound nature of high- without getting into trouble.” er education (Popenici & Kerr, 2017). The application of Some students miss easier fact-checking, like citing artificial intelligence to education has been the subject of sources in responses, which would enable the checking. academic research for many years. Jia et al. (2024) try to They miss better guidance on sources when searching for fill a gap in the current review of research on AI in science relevant professional literature. Although that, this is al- education (AISE) in the early stage of education by sys- ready possible with specific AI tools such as Perplexity. tematically reviewing existing research in this area. Dif- They would like to know “how to best obtain as accurate ferent researches are centred on various educational levels, information as possible from artificial intelligence since it fields, and contexts. For example, Xu and Ouyang (2022) often happens that it does not provide accurate information are discussing STEM education (science, technology, en- or does not ‘understand’ what they need and gives com- gineering, mathematics), Liang et al. (2021) language edu- pletely wrong answers.” Students also miss better AI tools cation, Drigas and Ioannidou (2013) special education, etc. for solving mathematical problems, better assistance in The field investigates learning wherever it occurs, in tradi- programming, and better and more accurate information. tional classrooms or workplaces, to support formal educa- They state “most AI tools cannot solve and explain more tion and lifelong learning. It brings together AI, which is challenging tasks. For example, ChatGPT often solves itself interdisciplinary, and the learning sciences (educa- computational tasks incorrectly.” tion, psychology, neuroscience, linguistics, sociology, and Many students see AI as a valuable asset for perform- anthropology) to promote the development of adaptive ing different tasks and know that AI doesn’t replace hu- learning environments and other AI tools that are flexible, mans. Many authors like Harding (2023), Hong (2024), inclusive, personalized, engaging, and effective. Brusilovsky (2024), Eisbach et al. (2024), Spillias et al. The article discusses using AI tools for educational pur- (2024), and others also support these findings. On the oth- poses from the perspective of students in higher education. er hand, some authors discuss the possibility and impli- Two hundred fifty-eight students from different faculties cations of artificial intelligence replacing humans. Roos of universities in Slovenia participated in the study. The (2023) highlights various professions where AI could re- findings indicate that while most students reported famil- place humans, including graphic designers, data analysts, iarity with AI concepts, their ability to articulate this un- programmers, and warehouse workers. Talmage-Rostron derstanding in their own words was limited. ChatGPT and (2024) discusses how AI impacts receptionists, account- Grammarly emerged as the most frequently used AI tools. ants, and salespeople jobs. AI systems are already per- The predominant use of AI tools among students was for forming tasks that range from managing reception duties information retrieval and idea generation. The study also to automating bookkeeping and sales processes, which unveiled some statistically important differences in using raises concerns about the future need for human workers AI tools based on gender and field of study. The study’s in these roles. ScienceDaily (2023) reports on the Univer- limitation is the sample size of 258 students from univer- sity of Waterloo’s findings that AI could replace humans in sity faculties in Slovenia. The study’s findings may also social science research. AI’s ability to process and analyze not be generalized to a larger population because they are 341 Organizacija, Volume 57 Research Papers Issue 4, November 2024 focused on students from a specific geographic location al. (2023). AI literacy in K-12: a systematic and may not capture the full diversity of AI tool usage and literature review. IJ STEM Ed 10(29). https://doi. familiarity among students in higher education globally. org/10.1186/s40594-023-00418-7 Further research with a more extensive and diverse sample Chandrasekar, S. (2023). Using Artificial Intelligence to and objective measures of AI tool usage and skills could Track Student Performance. AZoAi. provide a more comprehensive understanding of the use of Retrieved on April 24, 2024 from https://www.azoai. AI tools among students in higher education. com/article/Using-Artificial-Intelligence-to-Track- Student-Performance.aspx. References Chen, Y., Jensen, S., Albert, L.J., Gupta. S., & Lee, T. (2023). Artificial Intelligence (AI) Student Assis- tants in the Classroom: Designing Chatbots to Support Ahmad, S.F., Rahmat, M.K., Mubarik, M.S., Alam, M.M., Student Success. Inf Syst Front 25, 161–182. https:// & Hyder, S.I. (2021). Artificial Intelligence doi.org/10.1007/s10796-022-10291-4 and Its Role in Education. Sustainability, 13. https:// Coffey, L. (2023). Students Outrunning Faculty in AI Use. doi.org/10.3390/ su132212902 Tech & Innovation. Retrieved on May 25, Arora, M. (2021). Artificial Intelligence: New Pathways 2024 from https://www.insidehighered.com/news/ and Challenges in Higher Education. In tech-innovation/artificial-intelligence/2023/10/31/ S. Verma and P. Tomar (Eds.), Impact of AI Technologies most-students-outrunning-faculty-ai-use on Teaching, Learning, and Crompton, H. & Burke, D. (2023). Artificial intelligence in Research in Higher Education, (pp. 30-48). Igi Global higher education: the state of the field. Int Barrett, M., Branson, L., Carter, S., DeLeon, F., Ellis, J., J Educ Technol High Educ 20(22). https://doi. Gundlach, C., & Lee, D. (2019). Using org/10.1186/s41239-023-00392-8 Artificial Intelligence to Enhance Educational Dahri, N.A., Yahaya, N., Al-Rahmi, W.M. et al. (2024). Opportunities and Student Services in Higher Investigating AI-based academic support Education. Inquiry: The Journal of the Virginia acceptance and its impact on students’ performance in Community Colleges, 22 (1). Retrieved from https:// Malaysian and Pakistani higher commons.vccs.edu/inquiry/vol22/iss1/11 education institutions. Educ Inf Technol. https://doi. Baker, S. (2017). Which countries and universities are org/10.1007/s10639-024-12599-x leading on AI research? Times Higher Delcker, J., Heil, J., Ifenthaler, D. et al. (2024). First-year Education. Retrieved 26 May 2024 from https://www. students AI-competence as a predictor for t imeshighereducat ion.com/data-bi tes/which- intended and de facto use of AI-tools for supporting countries-and-universities-are-leading-ai-research learning processes in higher education. Int J Educ Basheer, K.C.S. (2024). Top 10 Countries Leading in AI Technol High Educ 21(18). https://doi.org/10.1186/ Research & Technology in 2024. Analytics s41239-024-00452-7 Vidhya. Retrieved 26 April, 2024 https://www. Drigas, A. S. & Ioannidou, R. (2013). A review on artificial analyticsvidhya.com/blog/2024/02/top-countries- intelligence in special leading-in-ai-research-technology/ education. Information Systems, E-Learning, and Bello, C. (2024). The best AI tools to power your academic Knowledge Management Research: 4th World Summit research. Euro news. Retrieved January on the Knowledge Society, WSKS 2011, Mykonos, 2024 from https://www.euronews.com/ Greece, September 21–23, 2011. Revised Selected next/2024/01/20/best-ai-tools-academic-research- Papers 4, 385–391. chatgpt-consensus-chatpdf-elicit-research-rabbit-scite Eisbach, S., Langer, M., & Hertel, G. (2023). Optimizing Bhullar, P.S., Joshi, M. & Chugh, R. (2024). ChatGPT in human-AI collaboration: Effects of higher education - a synthesis of the motivation and accuracy information in AI-supported literature and a future research agenda. Educ Inf decision-making, Computers in Human Behavior: Technol. https://doi.org/10.1007/s10639-024-12723-x Artificial Humans, 1(2), https://doi.org/10.1016/j. Brusilovsky, P. (2024). AI in Education, Learner Control, chbah.2023.100015. and Human-AI Collaboration. Int J Artif Farooq, E., Zaidi, E., & Shah, M. M. A. (2024). The Future Intell Educ 34, 122–135. https://doi.org/10.1007/ Classroom: Analyzing the Integration and s40593-023-00356-z Impact of Digital Technologies in Science Education. Campino, J.(2024). Unleashing the transformers: NLP Jurnal Penelitian Dan PengkajianIlmu Pendidikan: models detect AI writing in education. J. E-Saintika, 8(2), 280–318. doi: 10.36312/esaintika. Comput. Educ. https://doi.org/10.1007/s40692-024- v8i2.1957 00325-y Garrel, J. & Mayer, J. (2023). Artificial Intelligence in Casal-Otero, L., Catala, A., Fernández-Morante, C. et studies—use of ChatGPT and AI-based tools 342 Organizacija, Volume 57 Research Papers Issue 4, November 2024 among students in Germany. Humanities and Social Education. The74. Retrieved on April 24, 2024 from Sciences Communications, 10. 10.1057/s41599-023- https://www.the74million.org/article/shockwaves- 02304-7. innovations-how-nations-worldwide-are-dealing- Hanks, K. (2024). Which Country Is Best For AI with-ai-in-education/ Education? Global Leaders In Tech Learning. Lane, H.C. (2023). Commentary for the International Retrieved on March 25, 2024 https://inspireambitions. Journal of Artificial Intelligence in Education com/which-country-is-best-for-ai-education/ Special Issue on K-12 AI Education. Int J Artif Intell Harding, J., D’Alessandro, W., Laskowski, N.G. et Educ 33. https://doi.org/10.1007/s40593-023-00359-w al. (2023). AI language models cannot replace Laskowski, N. & Tucci, L. (2023). Artificial Intelligence. human research participants. AI & Soc. https://doi. Retrieved December 6, 2023, from org/10.1007/s00146-023-01725-x https://www.techtarget.com/searchenterpriseai/ Hashim, S., Omar, M. K., Jalil, H. A., & Sharef, N. M. definition/AI-Artificial-Intelligence (2022). Trends on Technologies and Artificial Liang, J., Hwang, G., Chen, M. A., & Darmawansah, D. Intelligence in Education for Personalized Learning: (2021). Roles and research foci of artificial Systematic Literature Review. International Journal intelligence in language education: An integrated of Academic Research in Progressive Education and bibliographic analysis and systematic review Development, 11(1), 884–903. approach. Interactive Learning Environments, 7, Hong, F. (2024). Group prioritarianism: why AI should not 4270–4296. replace humanity. Philos Stud. Lin, CC., Huang, A.Y.Q. & Lu, O.H.T. (2023). Artificial https://doi.org/10.1007/s11098-024-02189-5 intelligence in intelligent tutoring systems toward IBM (2023). What is a chatbot? Retrieved December 7, sustainable education: a systematic review. Smart 2023, from Learn. Environ. 10(41). https://doi.org/10.1186/ https://www.ibm.com/topics/chatbots s40561-023-00260-y Jaarsveld, S. & Lachmann T. (2017). Intelligence and Luckin, R. & Holmes, W. (2016). Intelligence Unleashed: creativity in problem solving: The importance An argument for AI in Education. London: of test features in cognition research. Front Psychol., Pearson 8. doi:10.3389/fpsyg.2017.00134 Malik, N. & Solanki, A. (2021). Simulation of Human Jensen, L.X., Buhl, A., Sharma, A. et al. (2024). Generative Brain: Artificial Intelligence – based learning. AI and higher education: a review of In S. Verma and P. Tomar (Eds.), Impact of AI claims from the first months of ChatGPT. High Educ. Technologies on Teaching, Learning, and https://doi.org/10.1007/s10734-024-01265-3 Research in Higher Education, (pp. 150-160). Igi Jia, F., Sun, D. & Looi, C. (2024). Artificial Intelligence in Global. Science Education (2013–2023): Naseer, F., Nasir Khan, M., Tahir, M., Addas, A., Aejaz Research Trends in Ten Years. J Sci Educ Technol 33, S.M.H. (2024). Integrating deep learning 94–117. https://doi.org/10.1007/s10956-023-10077-6 techniques for personalized learning pathways in higher Kaouni, M., Lakrami, F., & Ouidad, L. (2024). Integrating education, Heliyon 10(11). https://doi.org/10.1016/j. Artificial Intelligence and Natural heliyon.2024.e32628 Language Processing in E-Learning Platforms: A Popenici, S. & Kerr, S. (2017). Exploring the impact of Review of Opportunities and Limitations. 10.1109/ artificial intelligence on teaching and learning CiSt56084.2023.10409971. in higher education. Research and Practice in Kim, J. (2024). Leading teachers’ perspective on teacher- Technology Enhanced Learning, 12. AI collaboration in education. Educ Inf doi 10.1186/s41039-017-0062-8 Technol 29, 8693–8724. https://doi.org/10.1007/ Rensfeldt, A.B. & Rahm, L. (2023). Automating Teacher s10639-023-12109-5 Work? A History of the Politics of Automation and Korseberg, L. &Elken, M. (2024). Waiting for the Artificial Intelligence in Education. Postdigit Sci revolution: how higher education institutions Educ 5, 25–43. https://doi.org/10.1007/s42438-022- initially responded to ChatGPT. High Educ. https://doi. 00344-x org/10.1007/s10734-024-01256-4 Ramadhan, A., Warnars, H.L.H.S. & Razak, F.H.A. (2024). Korteling, J.E., van de Boer-Visschedijk, G.C., Combining intelligent tutoring systems Blankendaal, R.A.M., Boonekamp, R.C., & and gamification: a systematic literature review. Educ Eikelboom, A.R. (2021). Human- versus Artificial Inf Technol 29, 6753–6789. https://doi.org/10.1007/ Intelligence. Front. Artif. Intell, 4. doi: s10639-023-12092-x 10.3389/frai.2021.622364 Rawas, S. (2024). ChatGPT: Empowering lifelong learning Lake, R. (2023). Shockwaves & Innovations: How Nations in the digital age of higher Worldwide Are Dealing with AI in education. Educ Inf Technol 29, 6895–6908. https:// 343 Organizacija, Volume 57 Research Papers Issue 4, November 2024 doi.org/10.1007/s10639-023-12114-8 nexford.edu/insights/how-will-ai-affect-jobs Roos, D. (2023). 8 Jobs AI Will Replace and 8 It Won’t Wang, N. & Lester, J. (2023). K-12 Education in the Age (Yet). How stuff works. Retrieved May 16, of AI: A Call to Action for K-12 AI 2024, from https://electronics.howstuffworks.com/ Literacy. Int J Artif Intell Educ 33, 228–232. https:// future-tech/jobs-ai-will-replace.htm doi.org/10.1007/s40593-023-00358-x Rybina, G.V. & Grigoriev, A.A. (2023). Modern World Economic Forum (2024). Education and skills. The Architectures of Intelligent Tutoring Systems Based future of learning: How AI is on Integrated Expert Systems: Features of the Approach revolutionizing education 4.0. Retrieved May 6, 2024, to the Automated Formation of the Ontological Space from https://www.weforum.org/agenda/2024/04/ of Knowledge and Skills of Students. Pattern Recognit. future-learning-ai-revolutionizing-education-4-0/ Image Anal. 33, 491–497. https://doi.org/10.1134/ Wu, H., Li, S., Gao, Y. et al. (2024). Natural language S1054661823030409 processing in educational research: The Salido, V. (2023). Impact of AI-Powered Learning Tools evolution of research topics. Educ Inf Technol. https:// on Student Understanding and doi.org/10.1007/s10639-024-12764-2 Academic Performance. BAPS85: Introduction Xu, W. & Ouyang, F. (2022). The application of AI to Political Analysis and Research 10.13140/ technologies in STEM education: A systematic RG.2.2.17259.31521. review from 2011 to 2021. International Journal of Schroer, A. (2023). What is Artificial Intelligence? STEM Education, 9(1), 1–20. Retrieved December 6, 2023, from https://builtin.com/ Zheng, L., Niu, J., Zhong, L., & Gyasi, J. F. (2021). The artificial-intelligence effectiveness of artificial intelligence on ScienceDaily (2023). AI could replace humans in social learning achievement and learning perception: A me- science research. Retrieved July 6, 2024, from ta-analysis. Interactive Learning Environments, 31(9), h t t p s : / / w w w . s c i e n c e d a i l y . c o m / 5650–5664. https://doi-org.ezproxy.lib.ukm.si/10.108 releases/2023/06/230616161958.htm 0/10494820.2021.2015693 Seraydarian, L. (2021). How AI and Education Intersect. Retrieved December 7, 2023, from https://plat.ai/blog/ how-ai-and-education-intersect/ Eva Jereb is a professor in the Department of Personnel Singh, T. & Mishra, J. (2023). Learning with Artificial and Education Sciences at the Faculty of Organisational Intelligence Systems: Application, Challenges, Sciences, University of Maribor, Slovenia. Her main and Opportunities. In S. Verma and P. Tomar (Eds.), research interests are in higher education, e-learning, Impact of AI Technologies on Teaching, plagiarism, gamification in education, human resource Learning, and Research in Higher Education, (pp. development, self-management, personnel expert 236-253). Igi Global systems, and the phenomenon of telework. ORCID iD: Shen, L., Chen, I., Grey, A., & Su, A. (2021). Teaching and 0000-0003-1768-3787 Learning with Artificial Intelligence. In S. Verma and P. Tomar (Eds.), Impact of AI Technologies on Teaching, Learning, and Marko Urh obtained his Ph.D. in the field of Research in Higher Education, (pp. 73-99). Igi Global. Organizational sciences from the University of Spillias, S., Tuohy, P., Andreotta, M., Annand-Jones, R., Maribor. He is a senior lecturer in the Department of Boschetti, F., Cvitanovic, C. et al. (2024) Personnel and Information Sciences at the Faculty of Organisational Sciences, University of Maribor, Human-AI collaboration to identify literature for Slovenia. His main research interests are in higher evidence synthesis, Cell Reports Sustainability. https:// education, artificial intelligence, e-learning, human doi.org/10.1016/j.crsus.2024.100132. resource development, gamification and information Sultana, R. & Faruk, M. (2024). Does artificial intelligence systems. ORCID iD: 0000-0003-0278-2934 increase learners’ sustainability in higher education: insights from Bangladesh. J. of Data, Inf. and Manag. 6, 161–172. https://doi.org/10.1007/s42488-024-00121-4 Sumakul, D. T., Hamied, F. A., & Sukyadi, D. (2022). Artificial intelligence in EFL classrooms: Friend or foe? LEARN Journal: Language Education and Acquisition Research Network, 15(1), 232-256. Talmage-Rostron, M. (2024). How Will Artificial Intelligence Affect Jobs 2024-2030. Nexford Retrieved July 6, 2024, from University. https://www. 344 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Uporaba umetne inteligence med študenti v visokem šolstvu Ozadje/Namen: Umetna inteligenca (UI) ima vpliv na naše vsakdanje življenje, od poslovnih do družbenih področij, in v zadnjih letih vse bolj tudi na izobraževanje. Namen prispevka je dobiti vpogled v uporabo orodij umetne inteli- gence za izobraževalne namene iz perspektive študentov. Metode: Za raziskavo je bil uporabljen kvantitativni pristop z metodo anketiranja. Podatki so bili zbrani z e-anketo. Vprašalnik je vseboval zaprta vprašanja, ki so se nanašala na (i) splošne podatke, (ii) uporabo specifičnih orodij umetne inteligence, in odprta vprašanja (iii) o uporabi umetne inteligence za študijske namene. Za normalne in skoraj normalne porazdelitve odgovorov so bili uporabljeni parametrični t-test za neodvisne vzorce in enosmerna analiza variance (ANOVA). Rezultati: Študija, izvedena med 258 študenti v Sloveniji, razkriva splošno seznanjenost s koncepti UI, vendar z omejeno sposobnostjo artikulacije tega razumevanja. Priljubljena orodja UI, kot sta ChatGPT in Grammarly, se pre- težno uporabljajo za iskanje informacij in generiranje idej. Pokazale so se določene razlike v uporabi orodij UI glede na spol in področje študija. Zaključek: Hiter napredek UI pomembno spreminja visokošolsko izobraževanje. Integracija UI v izobraževanje spodbuja razvoj prilagodljivih, personaliziranih in inkluzivnih učnih okolij. Zaradi omejene velikosti vzorca in geograf- ske osredotočenosti študije so potrebne nadaljnje raziskave z bolj raznolikimi vzorci, da bi v celoti razumeli uporabo orodij UI v visokem šolstvu na globalni ravni. Ključne besede: Umetna inteligenca, Orodja UI, Visoko šolstvo, Študenti, Veščine 345 Organizacija, Volume 57 Research Papers Issue 4, November 2024 DOI: 10.2478/orga-2024-0025 Multilevel Investigation of Leadership Prototype Perception: Political Behavior in Relation to Effectiveness and Trust Aykut ARSLAN1, Serdar YENER2, Fatma KORKMAZ3, Murat Sakir EROGUL4 1 Department of Economics and Administrative Sciences, Piri Reis University, Istanbul, Turkey, aarslan@pirireis.edu.tr 2 Department of Business Management, Sinop University, Sinop, Turkey, syener@sinop.edu.tr 3 Faculty of Economics and Administrative Sciences, Yozgat Bozok University, Yozgat, Turkey, fatma.korkmaz@bozok.edu.tr 4 Department of Management, Adelphi University, Garden City, NY, USA, merogul@adelphi.edu Background/Purpose: Empirical evidence suggests that a leader’s political skills may act as a moderator that might decrease the effect of prototypicality’s impact on trust and on leadership effectiveness. The study investigated leading political skills as second-level regulatory variables in connection to leader effectiveness and trust with the purpose of testing a cross-level moderated mediation model within a traditionally collectivist culture like Turkey in efforts to contextualize and expand understanding of leadership prototype perception. Methods: A total of 442 service sector employees and 28 executive managers were interviewed. Two surveys in two separate time periods were conducted. Multilevel path analysis was used to evaluate the hypotheses. Results: The findings suggest that there is a strong and significant impact of leader prototypicality on leadership effectiveness and a direct impact on trust. Results indicate that leaders who are trusted by their subordinates are perceived as more effective in their leadership roles. Lower levels of trust from subordinates are negatively associ- ated with leader effectiveness. Conclusion: If leaders cannot provide their followers a contextualized sense of empowerment and development, they are perceived to not provide high levels of emotional trust. Therefore, there is greater need for multilevel con- textualized studies taking account of collective, two-sided, embedded experiences within groups. Keywords: Leader political behavior, Political skills, Leader effectiveness, Trust, Social identity theory, Turkiye 1 Introduction as persuasion and negotiation along with being frank, and the ability to use social relationships with skill are effec- The volatile and unpredictable conditions of today’s tive in achieving leadership goals and developing the trust complex world expose businesses to deal with increasing- in doing so (Ferris et al. 2005). Moreover, leaders who are ly difficult conditions. These challenging circumstances effective in understanding others within the organization, require effective leaders (Hasel 2013), and effective lead- and influencing them while aligning personal or organiza- ership in return requires the ability to create trust-based tional goals in accordance stand out as resourceful lead- relationships (Colquitt et al. 2007). Leadership skills such ers. This carries connotations of being political in one’s 1 Received: 24th November 2023; Accepted: 29th August 2024 346 Organizacija, Volume 57 Research Papers Issue 4, November 2024 leadership (Hochwarter et al. 2020). In some studies, it is such as persuasion, negotiation, coalition building, and reported that leaders who can successfully maintain po- leveraging social networks. With this said, to a larger ex- litical maneuvers appear to be effective leaders and these tent, many studies have focused on the perception of man- behaviors can sometimes result in a positive favoring the agers’ political behavior of employees (e.g., Mehmood et organization and its employees (Hochwarter 2012; Ellen al. 2019; Landells and Albrecht 2019; Hochwarter et al. III 2014; Kapoutsis and Thanos 2016; Hochwarter et al. 2020) without examining both the sense of trust in the lead- 2020). However, leadership literature has often concep- er and the role of the mediator. To fill this gap, a scale of tualized political leadership behavior as self-serving and trust was used to measure leader’s cognitive and emotional as productivity-hindering negative behaviors. These con- dimensions that are thought to be compatible with social ceptualizations out way the positive outcomes and have identity theory. Within the framework of the propositions caused negative perceptions (Hochwarter et al. 2020). Yet, of social identity theory, it is suggested that the leading organizations by nature are political arenas (Mintzberg, prototypes formed in the cognitive schemes of individu- 1985 as cited in Ferris et al. 2005), in which factors such als reflect the values and norms of the organization, which as sharing of scarce resources within the organization, de- can build trust in the leader (Hogg 2001; Barth-Farkas and grees of centralization and formalization are organic issues Vera 2019). Moreover, in a collectivist society like Turkey, that need to be led (Ellen III et al. 2013). Therefore, polit- leadership usually entails social cohesion and compliance; ical skills carry weight in achieving organizational goals in individualistic societies, it may give autonomy and per- and should be viewed as a resourceful skill rather than a sonal success top priority (Hofstede, 2011). Understanding negative expression (Hochwarter et al. 2020). these cultural subtleties is crucial as they affect the impres- Drawing from implicit leadership and social iden- sion of political actions and leadership models, therefore tity theory, our research assumes that positive thoughts affecting trust and efficacy. This study investigates these about the leader’s effectiveness can be formed through dynamics in the Turkish environment, offering insights the perception of trust in the leader. Leadership prototype that could have more general relevance for leadership in schemes are formed in the cognitive schemes of employ- cultural environments. The study is important in three ees that show the characteristics of leadership within their respects and provides the following contributions to the organizations (Giessner et al. 2009). It is possible to see the relevant literature: (1) Contributing to the research stream effect of leadership prototypes largely determined by this positive aspects of leader political skills, such as whether it cognitive affinity in the formation of a perception of trust increases confidence or not. Here, the perception of effec- in the leader (Barth-Farkas and Vera 2019). In addition, tive leadership may be due to the perception of the leader the perception of the leader prototype can be influenced by prototype rather than the leadership behavior (Mahon and the political skills of the leader, which can further increase Greenwald 2018). (2) Combining two points of view with- perceptions. Therefore, it may have a positive effect on the in the literature, leadership effectiveness and leadership confidence of the leader. In fact, the same regulatory vari- political behavior, that are seen as contradictory and stud- able model used in this study had previously been used by ied as serving opposite sides. A lack of multilevel studies, Giessner et al. (2009). Similarly, the relationship between as argued by Mahon and Greenwald (2018) state that the leadership prototypicality and leadership effectiveness has perception of the leading prototype if well understood by been the subject of previous research. organizations, can have many positive contributions for In this study, mutual data from both employees and the employees, from the quality of leader-member interaction leaders of these employees is collected and examined. The to being good at work. Claiming that this may be overcome sample consists of service sector employees and managers with more studies exploring leadership effectiveness and in Ankara, the capital of Turkey, and the surrounding prov- political leadership skills together as a resource for leader- inces and districts. Unlike the Giessner et al. (2009) study, ship development. (3) Clearing the vagueness in relation to we have modeled leading political skills as second-level the argument regarding the perception of leadership within regulatory variables and created a cross-level moderated a group by using a multilevel model that permits evaluat- mediation model. We tested whether the model also works ing within- and between-level variance. Currently, many in a collectivist culture like Turkey (see Hofstede 1980, studies within organizational sciences have evaluated the 2011), in which a leader’s political skills may act as a mod- effects of political constructs at the individual level (Ferris erator that might decrease the effect of prototypicality’s et al. 2019), whereas the use of aggregation and multilevel impact on trust and on leadership effectiveness. Within the modeling will allow the means to better evaluate the col- framework of organizational management, we conceptual- lective, dyadic, nested, and within-person effects that have ize political behavior used to describe the behavior of indi- been sparse within the literature. viduals in order to have, develop and use power and other In the following section, we outline a brief review of necessary resources to achieve desired results in uncertain the literature on social identity theory, leader prototype, or conflictive situations. We acknowledge that political the role of trust in the leader, and political leader behavior leaders often involve tactics outside of formal processes, and skills as we develop the theoretical foundation. Next, 347 Organizacija, Volume 57 Research Papers Issue 4, November 2024 we describe the methodology used, followed by the find- followers develop leadership behavior schemes specific to ings. In the discussion section we discuss the theoretical their group through their past experiences and interactions implications and contributions of our findings. Lastly, within the group (Ridgeway 2001). It is within the frame- we wrap up with a brief conclusions section and describe work of these schemes that employees act according to the some limitations of our study. norms of the group they are in, they create expectations about the behavior of leaders (Hogg 2001; Ibarra et al. 2 Literature Review and Hypotheses 2010). However, social identity plays a guiding role in the behavior of followers and also leads the organization to prioritize its interests (van Knippenberg 2011). In this con- 2.1 Social Identity Theory text, this research model has been formed by considering that the leaders who behave in accordance with the norms Leadership is about how some individuals or cliques of the group will gain the trust of the employees and will have disproportionate power and influence to set an be supported and thus carry out their duties effectively. Ac- agenda, define identity, and mobilize people to achieve cording to this theoretical model, it is suggested that the collective goals (Hogg 2001). However, activity and per- perception of the leader prototype shaped in the followers formance in social contexts such as organizations can in- can be considered as “one of us” (Lord and Hall 2003:57) crease based on the dynamic interaction of the leader and and this interaction can create a sense of trust in the lead- its members (van Knippenberg 2011). Only personal char- er over time. Through this sense of trust, it is estimated acteristics of leaders or only social or situational factors that the leader’s perceptions of his effectiveness would in- alone will be inadequate in achieving effectiveness (Ha- crease. In addition, in this relationship, it is thought that levy et al. 2011). Tajfel (1972) introduced the idea of social the leader can make a positive contribution to the effect of identity to theorize how people conceptualize themselves the leader prototype by showing political behavior skills in in intergroup contexts, how a system of social categori- accordance with the norms of the group and thus increase zations “creates and defines an individual’s own place in the level of trust. society” (p. 293). He defined social identity as “the indi- vidual’s knowledge that he belongs to certain social groups 2.2 Leader Prototype together with some emotional and value significance to him of this group membership” (Tajfel 1972: 292). When examining the determinants of the leader’s effectiveness, Today, when we look at the studies that deal with the it has been revealed in many studies that social identity dynamics of leadership, we see that the leader emphasiz- formation occurs through multilateral and dynamic pro- es interaction within the group more (van Knippenberg cesses (Hogg 2001), based on social categorization and 2011). It is revealed in studies that successful leaders have prototypes within groups. Within the framework of SIT, it high interaction within the group, and that high interac- is suggested that leadership is shaped by dynamic interac- tion shows compliance with group norms (Mahon and tions between leaders and followers (Hogg 2001). Due to Greenwald 2018; Barth-Farkas and Vera 2019). There- the direct and indirect influence of the group on the indi- fore, success of leaders in adapting to group norms, may vidual, it is also reported that individuals adapt cognitively bring about effectiveness of the leader and organization and behaviorally to the dominant identity characteristics (Barth-Farkas and Vera 2019). formed in the group (van Knippenberg 2011). This process Indeed, leadership behaviors in accordance with group can occur as a result of the strengthening process of leader norms form the basis of the ideal leader prototype percep- over time, depending on their personal characteristics or as tion in employees and are used as the main criterion for a result of the proactive behavior of the leader. However, positioning the leader (Jacquart and Antonakis 2015). The the influence of the leader alone is not enough in this inter- leader prototype reflects the characteristics of the schemes action. Influence must also mobilize followers and begin formed in the followers, and develops in accordance with to support the status and power of their leaders (Fiske and the norms of the group (organization). (van Knippenberg Depret 1996). The leader’s creativity and particular behav- 2011). Leaders who act in accordance with the norms of iors, and their acceptance by followers’ affect how much the organization and the schemes of the followers form the the leader can push the boundaries (idiosyncrasy credit) ideal leader prototype, which shapes the attitudes and be- depending on the group norms (Hogg 2001). haviors of the employees towards the leader and the organ- Within the extant literature, it can be seen that this in- ization (Giessner et al. 2009). Leaders who reflect impor- teractive process is initiated by the leader’s charismatic, tant values and norms for the organization in their attitudes innovative perspective with vision and mission. Hence, and behaviors emphasize that these values and norms are followers are also influenced by the leader, strengthening even more important and create awareness in employees in this process and therefore the status of the leader (Guillen this direction (Mayer et al. 1995). When employees see the et al. 2015, Greenberg et al. 2007). It is also claimed that norms and values of relatable social identity in the leader’s 348 Organizacija, Volume 57 Research Papers Issue 4, November 2024 behavior, identification process with the leader and the or- ly, when leaders are seen as group members, i.e. as “one of ganization can take place (Dirk and Ferrin 2002). In this us” (Steffens et al. 2018), they are “trusted as the most re- process, it may also be effective for the leader to set goals liable source of identity-related information” (Barreto and and objectives that may coincide with the group values and Hogg 2017: 42). Consequently, followers endorse, support norms. This process enables the leader to convey his or and allow leaders to be effective” (p.42). Furthermore, her vision to their followers and to connect emotionally. they believe that prototypical leaders are usually seen as Ultimately, the perception that “we are one” and that “we highly trustworthy (Barreto and Hogg 2017; Steffens et al. can only succeed if we are one” begins to form in the fol- 2018). However, due to its referral to a follower perception lowers. This can make followers see the leader as a team (e.g. personality, leadership style), effectiveness could not player, strengthening their sense of belonging, attachment be directly related to prototypicality. The concept of group and trust over time. prototypicality is defined as having a very close psycho- The general framework of the leader schemes con- logical relationship with trust than effectiveness (Barreto sists of the leading figures who are loved and connected and Hogg 2017), which implies an indirect relationship. (Mayseless 2010). These cognitive descriptions can also From the perspective of the classifications of Hofstede create a framework for an effective leader, either directly (1980, 2011) and Schwartz (2006) it is unclear whether or indirectly. A recent meta-analysis revealed that proto- trust in individualistic societies can produce the same re- typicality accounted for 24% of variance in leader eval- sults in a collectivist society’s (Shavitt et al. 2008, 2011). uation (Barreto and Hogg, 2017). The concept of leader This issue has not been the subject of research. The un- effectiveness or effective leader is defined differently in the deniable effect of the perception of trust in the leader on literature. For example, de Vries et al. (2015) refers to con- the performance of organizations (Morgeson et al. 2010) tingency and defines leader’s effectiveness as being able to requires more studies on trust in the leader in societies with adapt easily to any situation. Barreto and Hogg (2017: 45). different cultural characteristics. Especially considering measured the effectiveness with observing “how success- the importance of trust within organizations and the wid- fully they achieve group goals” in their meta-analysis. On er business context. New studies are needed in collectivist the other hand, Giessner et al. (2009) simply defines it as societies where distinguishing “us” versus “them” has the organizational success/failure, while in other studies (Fein effect of identifying others who can be trusted or not. This et al. 2020) it is defined as the success of motivating fol- may be particularly relevant in the context of collective lowers to achieve organizational goals and managing the societies where social and collective identity is essential expectations and needs of followers while doing so. Since (Hogg 2001; Hogg et al. 2004). both the first and fourth definitions refer to a relational pro- Within frames of literature review; cess, they are suitable for our research model in this article. H2a: Trust in leader is a mediator between leader pro- In alignment, Stevens et al. (2019) argues that the leader totypicality and leadership effectiveness. effectiveness is caused by the sense of identity shared by H2b: Trust in leader is a moderator between leader its group members. prototypicality and leadership effectiveness. Within frames of findings presented beyond; H1: Leader prototypicality has a direct impact on 2.4 Leader Political Behavior and Skills leadership effectiveness. Political behavior is a form of informal influence 2.3 The Mediating Role of Trust in the that is a foundational element of leadership (Ferris and Leader Hochwarter 2011). Politically skilled leaders are able to adapt themselves to the environment through their so- The role of trust in leadership literature is referenced cial intelligence, allowing them to read interpersonal thoroughly throughout the literature (Goodwin et al. 2011) relationships with the power of intuition, and influence and emphasized repeatedly (Liden et al., 2015). Trust in those around them, in return establish new relationships leadership is studied as an outcome, a moderator, and as (Özdemir and Gören 2015). Hochwarter et al. (2020) re- a mediator variable in research models within the litera- veals that organizational policy perceptions can affect the ture. Trust is accepted as one of the most important aspects stress, performance and health status of employees in the of leadership qualities (Platow et al. 2003). As claimed workplace requiring a different perspective on the subject. by Podsakoff et al. (1996), trust seems to be an important Within the framework of the extant literature, it can be said variable that has the power to mediate the effectiveness that the concepts of organizational politics, political skills of transformational leadership. Likewise, our suggestion is and political behavior interact with each other and develop that it is a significant mediator of leader effectiveness. The together (Ferris and Treadway 2012). When we look at this model is based on the triggering effect of leader behavior interaction from the point of view of the leader, one of the on followers’ perception of trust and respect to their leaders factors affecting the political ability of the leader is seen as and in return a feeling of enhanced motivation. Additional- organizational policy perceptions (Chang et al. 2009). This 349 Organizacija, Volume 57 Research Papers Issue 4, November 2024 relationship from the perspective of employees reveals 442 employee and 28 manager surveys were used and in- that organizational policy perceptions may vary according cluded in the analysis. All the scale items were measured to the value perceptions of the societies they live in (Ziben- using a five-point Likert scale from 1 = “strongly disagree” berg 2017). The fact that the value perceptions of societies to 5 = “strongly agree.” Whereas political skills were in terms of employees affect the behaviors of employees, measured at the leader level, leadership prototypicality, such as emotional intelligence and extra-role behavior trust in leaders, and leadership effectiveness were assessed (Lvina et al. 2012). Hence, facilitating the political skills using several employees’ ratings of the same manager. of the leaders to be shaped and developed. In addition, To account for the nested nature of the data (i.e., sub- considering that employees’ perceptions are constructed ordinates were nested inside supervisors), we utilized according to cultural codes (Kastanakis and Voyer 2014), multilevel path analysis to evaluate the hypotheses. The one might think that culture can shape both organization- variance of a Level-1 variable is decomposed within a al policy and political skill. Cultural codes, and therefore component (within-group variance). and a between com- perceptions of value, may lead to the leader’s development ponent (between-group variance). The definition of struc- of political behavior and skills through employees’ percep- tural linkages may be used to describe the relationships tions of policy. between these variance components at each level (Muthén Although the perception of organizational policy in and Asparouhov 2009). different cultures has been compared with the concept of The intraclass correlation coefficient (ICC). is com- ethics (Zibenberg 2017), as stated above, there is a need monly used in multilevel research to assess the extent of for cultural studies in which organizational policy, organi- variability between groups. In this study, the ICC was zational skills and political behaviors are examined togeth- calculated using variance components derived from a one- er and contextualized from the point of view of both the way random effect ANOVA (LeBreton et al. 2023). Specif- leader and the employee’s social and cultural perceptions. ically, the ICC(1) statistic was employed to understand the Within frames of literature review; discrimination between groups based on the aggregation H3a: Leader political skill act as a moderator and of scores using unit-level means. For interpreting the ICC decreases the impact of leader prototypicality on trust in values, a significant ICC(1) value falling within the range leader. of 0.05-0.20 is considered noteworthy (Bliese 2000). Ad- H3b: Leader political skill act as a moderator and de- ditionally, the ICC(2) statistic, which also relies on var- creases the impact of leader prototypicality on leadership iance components obtained from the one-way random effectiveness. effects ANOVA, provided insight into the discrimination between groups based on means (LeBreton et al. 2023). 3 Methods A recommended cut-off value for ICC(2) is 0.60 (Glick 1985; Schneider et al. 1998). To assess within-group con- cordance, rwg(j) values were utilized (James et al. 1984). 3.1 Procedures Specifically, a one-way analysis of variance (ANOVA) was conducted to determine if responses from multiple The participants of the study consist of service sector direct reports converged and could be aggregated. The employees and managers in Ankara, the capital of Turkey, ANOVA yielded a significant result, F(27, 393) = 6.89, p and the surrounding provinces and districts. The HR man- < 0.000. The calculated ICC(1) and ICC(2) values were agers of the organizations. Two of these were mid-sized 0.28 and 0.85, respectively, indicating moderate and high and three were large sized organizations. They were par- levels of discrimination between groups. Furthermore, the ticularly chosen because of their team structure working average Rwg (multi-item scale) was found to be 0.86, sug- environment. Each team was run by a team leader and gesting acceptable within-group concordance (Klein and members vary from 4 to 42. They were contacted and the Kozlowski 2000). necessary permissions were obtained. The questionnaires Multilevel path analysis was utilized, as suggested by were sent to the contacted HR managers. The researchers Preacher et al. (2010) due to it being more suited for as- then visited the institutions one by one in order to admin- sessing multilevel mediations in comparison to the hierar- ister and collect the filled-out questionnaires by hand. The chical linear modeling technique. It is also believed that it surveys were conducted in two separate time periods, the extends Baron and Kenny’s (1986) multi-step regressions first on January of 2022 and the second on the month of to the multilevel situation. Sun et al. (2012:60). put forth February of 2022. A survey of managers (Leader Political the strengths of multilevel path analysis as follows; “a Skills) and employees (Trust in leader) were conducted in multilevel path model can (1) avoid the potential problem the first visit. In the second, the remaining two question- of conflating between-group and within-group relation- naires were filled. A total of 445 employees and 32 execu- ships, (2) directly estimate indirect effects and the multiple tive surveys were filled in. However, as a result of the sur- paths that are components of these indirect effects, and (3) veys that were removed for incomplete or other reasons, provide fit indices for the overall model”. Thus, we em- 350 Organizacija, Volume 57 Research Papers Issue 4, November 2024 ployed SPSS 23 and MLMED - BETA VERSION 2 for Both coefficient alpha (α = .89). and composite reliability multilevel path analysis. estimates (CR = .93) for the higher order construct were adequate. The one-factor model had acceptable fit indices 3.2 Measures (X2=0.002, df=1; RMSEA=.00; CFI=.1; TLI=.1). Trust in Leader Scale: The scale consisted of two fac- Political Skill Inventory: Subordinates’ perceptions tors and ten items. Originally it was developed by McAl- of their leaders’ political skill were measured by this lister, 1995 and adapted to Turkish by Ari (2003). Items 18 item-inventory, developed by Ferris and his friends assessed Cognition-based (e.g. “This person approaches (2005). Items assessed leaders’ social astuteness (e.g. “He/ his/her job with professionalism and dedication”) and Af- she has good intuition or “savvy” about how to present fect-based (e.g. “We would both feel a sense of loss if one him/herself to others”), perceived interpersonal influence of us was transferred and could no longer work togeth- (e.g., “He/she is good at getting people to like him/her”), er”). The reliability of this score was high, with a Cron- apparent sincerity (e.g., “He/she tries to show a genuine in- bach’s Alpha of .96 so is the composite reliability estimate terest in other people”), and networking ability (e.g., “He/ (CR=.94). Additionally, cognitive-based trust’s reliabil- she is good at using his/her connections and networks to ity score was .92 and Affect-based trust was .93. Lastly, make things happen at work”). Both coefficient alpha (α both factors’ CR estimates (CR=.93 and CR=.93) were = .92) and composite reliability estimates (CR = .82) for adequate. The two-factor model had acceptable fit indices the higher order construct were adequate. The four-factor (X2=95.7, df=32; RMSEA=.07; CFI=.98; TLI=.97). model had acceptable fit indices (X2=386, df=129; RM- Leadership Effectiveness Scale: Incorporating nine SEA=.07; CFI=.93; TLI=.91). items, the scale was developed by Ng & Chan (2008) and Leader Group Prototypicality Scale: The participants adapted into Turkish by the authors. The scale assesses the were asked to rate their agreement on six items developed leadership effectiveness through items such as “planning by van Knippenberg and van Knippenberg (2005). It was ability” and “setting direction”. Both coefficient alpha (α adapted to Turkish by Ömür (2018). Items assessed lead- = .97) and composite reliability estimates (CR = .96) for er prototypicality through statements such as “The team the higher order construct were adequate. The one-factor leader shares a lot of similarities with the members of my model had acceptable fit indices (X2=66.8, df=23; RM- team”. To determine the composite leader group prototyp- SEA=.07; CFI=.99; TLI=.98). icality score, the responses to each item were summed. Table 1: Confirmatory factor analysis of key variables in the study Factor structure model X2(df) X2/df GFI CFI SRMR RM- SEA Δχ2(Δdf) Three factor (hypothesized): Leader group prototypicality, leader effec- 326.4(146) 2.235 0.923 0.928 0.024 0.054 tiveness, trust in leader Two factor Model 1 (alternative): Leader group prototypicality and leader effective- 1122.7(187) 6.003 0.808 0.901 0.060 0.109 796.3(41) ness constrained as one factor Model 2 (alternative): Leader effectiveness and trust in leader 1755.6(188) 9.338 0.577 0.833 0.067 0.141 1429.2(42) constrained as one factor Model 3 (alternative): Leader group prototypicality and trust in leader 1156.7(188) 6.152 0.798 0.897 0.045 0.111 830.3(42) constrained as one factor One factor (alternative): All three scales together as one factor 2167.6(189) 11.469 0.554 0.790 0.074 0.158 1841.2(43) * Note: N = 420, All χ2 and Δχ2 values are significant at p < 0.05, χ2 =Chi-Square, df=Degree of Freedom, GFI=Goodness of Fit Index, CFI=Comparative Fit Index, SRMR= Standardized Root Mean Square Residual, RMSEA=Root Mean Square Error of Approximation 351 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Controls: We accounted for several demographic var- school education; another 11% had a vocational school di- iables related to employees. Gender orientation was cate- ploma; 57% got a bachelor’s degree and 11% had a mas- gorized as 0 for Male and 1 for Female. Additionally, we ter’s degree. controlled for employees’ age, education level, and tenure (number of years spent working with supervisors). 4.1 Confirmatory Factor Analyses 4 Findings To begin, we carried out an omnibus confirmatory factor analysis (CFA) to confirm that the three employ- From the 420 subjects, 237 were women and 183 were ee-reported variables were distinct: Leader group pro- men. Their age cohorts were as follows: 18% varied 18-25; totypicality, trust in leader, and leadership effectiveness. 25% varied 26-30; 31% varied 31-40; 18% varied 41-50; Individually, we compared the CFA models. The three-fac- 9% varied > 51. Additionally, 11% completed secondary tor model performed better than the null model and the education and 38% high school education; 21% obtained one-factor model (see Table 1). The comparative fit index a vocational school diploma; 24% got a bachelor degree; (CFI) and root-mean-square error of approximation (RM- and 6% had a master’s degree. On average, they have SEA). of the three-factor model were also higher than worked in their current organizations for about 17 years. those of the one-factor model (CFI=.92, RMSEA=.054). The average age of the 28 managers was 43 years with As a result, we treated these three measures as distinct con- a tenure of 25 years; There were 5 women and 23 men. structs as we proceeded with our analyses. Approximately 21% of the sample of managers had a high Table 2: Means, standard deviations, and correlations between study variables Mean S.D. 1 2 3 4 5 6 7 8 9 Individual Level 1.Gender 0.436 0.496 1 2.Age 2.75 1.20 -143** 1 3.Education 2.76 1.12 -0.028 0.079 1 4.Tenure w. supervisor 2.77 1.47 -0.156** 0.630*** 0.292*** 1 5. Organizational tenure 14.4 6.75 0.063 0.070 0.027 -0.002 1 6. Trust in leader 3.81 0.974 -0.041 -0.059 -0.120* -0.165*** 0.053 (.96) 7. Group Prototypicality 3.45 1.12 -0.017 -0.019 -0.045 -0.120* 0.108* 0.593*** (.89) 8. Leadership effectiveness 3.84 1.12 -0.025 -0.054 -0.087 -0.160*** 0.014 0.790*** 0.515*** (.97) Group Level 9. Leader (.92) political skill 4.29 0.317 0.058 -0.158** -0.340*** -0.204*** 0.020 -0.009 -0.041 -0.003 (1) Individual level N=420; group level N=28 (2) Coefficient alphas are listed in parentheses along the diagonal (3) Gender is coded as 0=female; 1=male (4) Age is coded as “1” = 18-25, “2” = 26-30, “3” = 31-40, “4” = 41-50, “5” = above 51. (5) Education is coded as “1” = primary, “2”= High school, “3” = associate degree, “4” = bachelor’s degree, “5” = master’s degree. (6) Note. * p < .05, ** p < .01, *** p < .001 352 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 3: The Results of Multi-level Regression Analyses Model 1: Multilevel Regression Model for Trust in Leader 95% Confidence Interval for Mean Level 1 Estimate Lower Bound Upper Bound constant 4,55*** (0,53) 3,46 5,65 Leader Political Skill X Leader Prototypically -0,24* (0,12) -0,47 -0,01 Leader Prototypically 0,46*** (0,03) 0,40 0,53 Gender -0,07 (0,07) -0,21 0,08 Age -0,01 (0,04) -0,09 0,06 Education level (0,05) -0,23 -0,04 Level 2 95% Confidence Interval for Mean Estimate Lower Bound Upper Bound Leader Political Skill -0,23 (0,28) -0,80 0,35 Gender -0,30 (0,53) -1,40 0,79 Age -0,15 (0,15) -0,46 0,15 Tenure with manager -0,2 (0,12) -0,45 0,05 Organizational tenure 0,03 (0,05) -0,07 0,13 Education level 0,08 (0,15) -0,22 0,38 Model 2 Multilevel Regression Model for Leader effectiveness Level 1 95% Confidence Interval for Mean Estimate Lower Bound Upper Bound constant -0,38 (0,52) -1,49 0,74 Leader Political Skill X Leader effectiveness -0,23* (0,10) -0,43 -0,03 Leader Prototypically 0,07 (0,04) -0,01 0,14 Trust in leader 0,68*** (0,04) 0,59 0,76 Gender 0,01 (0,06) -0,11 0,13 Age -0,04 (0,03) -0,10 0,03 Tenure with manager -0,06 (0,03) -0,11 0,00 353 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 3: The Results of Multi-level Regression Analyses (Continues) Estimate Lower Bound Upper Bound Organizational tenure 0,00 (0,00) -0,01 0,01 Education level -0,05 (0,04) -0,13 0,03 Level 2 95% Confidence Interval for Mean Estimate Lower Bound Upper Bound Leader Political Skill 0,15 (0,15) -0,16 0,45 Trust in leader 1,04*** (0,10) 0,83 1,25 Gender 0,02 (0,27) -0,53 0,57 Age 0,03 (0,07) -0,12 0,19 Tenure with manager 0,02 (0,06) -0,12 0,15 Organizational tenure 0,00 (0,03) -0,05 0,05 Education level 0,05 (0,07) -0,10 0,20 -2LL 1766,412 * p < 0.05, ** p < 0.01, *** p < 0.001 n= 420;n=28 Estimates are non-standardized Standard errors in parenthesis 4.2 Common Method Bias (r=.79, p<0.001), group prototypicality was positively correlated to leadership effectiveness (r=.52, p<0.001). In order to mitigate the potential influence of common At the group level, leader political skill had no significant method bias, data for the variables in this study were col- relationship with the other variables. However, it was neg- lected from two different sources. This approach helps to atively related to subordinates’ age (r=-.16, p<0.01), edu- address the issue of common source variance (Podsakoff et cation level (r=-.34, p<0.001), and tenure with supervisor al., 2003). The results of the confirmatory factor analysis (r=-.20, p<0.001). Organizational tenure was only corre- provided evidence that the four-factor measurement model lated to group prototypicality (r=.11, p<0.05). (χ2 = 2167.6, df = 189, GFI = 0.55, CFI = 0.79, RMSEA We asserted that leader prototypicality affects leader = 0.16) outperformed the one-factor measurement model. effectiveness in our first hypothesis, stating ‘leader proto- This indicates that the presence of common source var- typicality has a direct impact on leadership effectiveness’. iance was not a significant concern in the present study A regression analysis was conducted to examine the rela- (Podsakoff et al., 2003). tionship between leader prototypicality and leader effec- tiveness. The standardized coefficient for leader prototypi- cality is 0.515, indicating that for a one-standard-deviation 4.3 Hypothesis Testing increase in leader prototypicality, the dependent variable, leader effectiveness, increases by 0.515 standard devia- Table 2 below exhibits means, standard deviations and tions. This finding suggests a strong and significant im- intercorrelations of the study variables. At the individual pact, thus supporting our first hypothesis. level, trust in leader was positively related to group pro- Relooking at Table 3 for our second hypothesis, we totypicality (r=.59, p<0.001) and leadership effectiveness claimed a mediation effect of trust in the leader at Level 1. 354 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Before reaching a conclusion, we checked the sub-hypoth- 4.4 Moderations eses. For H2a, we posit that leader prototypicality has a di- rect impact on trust in the leader. This assertion is support- In our first model, at Level 2, we conducted the first ed (b = 0.46, SE = 0.03, p < 0.000, 95% CI [0.40, 0.53]). interaction analysis to examine how leader political skill Our second sub-hypothesis, H2b, posits that trust in moderates the relationship between leader prototypicality the leader has a direct impact on leader effectiveness. The and trust in the leader. The interaction between leader po- results of this assertion is presented in Model 2. We find litical skill and leader prototypicality was negatively as- support for this hypothesis as well (b = 0.67, SE = 0.04, p sociated with trust in the leader (b = -0.24, SE = 0.12, p < < 0.000, 95% CI [0.59, 0.76]). Leaders who are trusted by 0.05, 95% CI [-0.47, -0.01]). This suggests that when lead- their subordinates are perceived as more effective in their ers with high political skill were also perceived as more leadership roles. prototypical, they tended to have lower levels of trust from The last checkpoint is to examine if the prediction their subordinates. Thus, we find support for H3a. of the independent variable on the dependent variable is As seen in Figure 1, when the employees perceive low non-significant. Again, in Model 2, our analysis reveals a level of political skill, and higher levels of leader prototyp- non-significant effect for this (b = 0.07, SE = 0.04, n.s., icality, trust in their leader increases. 95% CI [-0.01, 0.14]). Thus, we can conclude that a me- In the second model, at Level 2, the analysis revealed diation effect exists at Level 1, and H2 is also supported. that the interaction between leader’s political skill and Our second model is also a multilevel regression anal- leader effectiveness was negatively associated with lead- ysis that examines the factors influencing leader effective- er effectiveness (b = -0.23, SE = 0.10, p < 0.05, 95% CI ness. The Level 1 results indicated that several variables [-0.43, -0.03]). This suggests that leaders with high polit- significantly predicted leader effectiveness. At Level 1, the ical skill who also demonstrated effectiveness had lower constant term was found to be -0.38 (SE = 0.52, 95% CI overall levels of leader effectiveness (See Figure 2). This [-1.49, 0.74]), suggesting that the average level of leader supports our H3b. effectiveness was -0.38. However, none of the Level 1 pre- The -2LL for the model was 1766.412, indicating a dictors, including leader prototypicality, trust in the leader, good fit to the data. Please note that the all estimates pro- gender, age, tenure with manager, organizational tenure, vided, except for the first analysis of hypothesis one, are and education level, had statistically significant associa- non-standardized. tions with leader effectiveness. Figure 1: First Moderation Results 355 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Figure 2: Second Moderation Results 5 Discussion found to also have a direct impact on trust in the leader and their perceived leader effectiveness. Thus, demon- The paper has examined leading political skills within strating that leaders trusted by their subordinates are also the framework of social identity theory, in which internal perceived as more effective in their leadership roles. To interactions within the organization may impact the atti- add on, leaders with high political skills are perceived to tudes and behaviors of leaders and the perception of the be more prototypical, with lower levels of trust from their leader prototype of the employees. The study employed subordinates. Their leader effectiveness also was negative- second-level regulatory variables and created a cross-lev- ly associated with leader effectiveness. el moderated mediation model to test the model within a The contributions of the study are threefold. Firstly, collectivist culture like Turkey. In general, different leader contributing to the leadership literature positive aspects of skills and perception of the leader prototype may occur in leader political skills in comparison to a large extant of different organizations. Therefore, with this in mind, data literature presenting the negative aspects (Hochwarter et obtained in this study was collected from the employees of al. 2020) of political behavior of leaders. More particu- different organizations, and samples were compared with larly, the perception of effective leadership is revealed as the data obtained from the leaders of the same employees. a perception of the leader prototype rather than to the be- In doing so, the study measured subordinates’ perceptions havior of the leader (Mahon and Greenwald 2018). The of their leader’s political skills (Giessner et al. 2009) in study revealed that leaders adopted and aligned their lead- order to assess their social astuteness in terms of how they ership behavior to the organizational culture in order to perceived their leader’s interpersonal influence, sincerity influence their followers and direct them to accomplish the and networking ability. In addition, the study examined goals and objectives of the organization. Although Stef- ratings in relation to the leader’s prototypicality, as well as fens et al. (2021) draws attention to the difference between both cognition-based (i.e. professionalism and dedication group prototypicality and leader prototypicality in that the to job) and affect-based (i.e. sense of loss of transferred or notion of leader prototypicality is used in works focusing could no longer work together). trust in the leader. Lastly, on implicit leadership theories, this study took a different leadership effectiveness was measured through perceived perspective. The study confirmed that at the individual ‘planning ability’ and ‘setting direction’. The combining level trust in leader was positively related to group proto- of leadership effectiveness and leadership political be- typicality and leadership effectiveness, indicating that the havior within the study showed that no matter how much average group prototype represented the perceived char- emotional relationship is established, if leaders who do acteristics or qualities that are most commonly associated not provide their followers a sense of empowerment and with leadership within a particular group or organization. development, they do not produce a high level of emo- Moreover, it reflected the collective perception of what tional trust. More precisely, a strong and significant impact constitutes effective leadership based on the experiences, of leader prototypicality on leadership effectiveness was norms, and values prevalent within the group (Shavitt et al. 356 Organizacija, Volume 57 Research Papers Issue 4, November 2024 2008, 2011). The study conjectures that the average group Dick and Kerschreiter 2016). The validity of social iden- prototype may emerge from observing and generalizing tity theory is proven in Turkish society, in which leader’s the behaviors and traits displayed by leaders who have political skills compatible with norms and cultures are been successful in that particular context. Moreover, it functional. Therefore, it is important to note that cultural represents the common expectations and standards held by influences on the perception of leader political behavior group members regarding leadership, serving as a bench- are complex and multifaceted. Cultural dimensions in- mark for evaluating leaders within the group. Hence, the teract with individual differences, organizational factors, average group prototype of leadership is perhaps shaped and situational factors, creating a dynamic context for by various factors such as organizational culture, historical interpreting and evaluating leader political behavior. Un- leadership practices, and the shared beliefs and values of derstanding these cultural influences can help leaders and group members. organizations navigate cross-cultural contexts and adapt Secondly, the study combines leadership effectiveness their leadership behaviors accordingly. When proposi- and leadership political behavior in cooperation to one an- tions of social identity theory are examined in the light of other rather than as contradictory and opposite sides of or- Zibengerg’s (2017) studies, it can be thought that organi- ganizational and leadership collaboration. In doing so, we zational politics, political skills and perceptions of leader demonstrate the positive contributions in confirmation to prototypes can be influenced not only by organizational Mahon and Greenwald (2018). In contrast to a large num- culture, but also by the characteristics of society. ber of studies within the extant literature focused solely on Culture plays a significant role in shaping the percep- the political behavior in respect to the perception of man- tion of leader political behavior within an organization. agers’ without examining other relations (e.g. Mehmood These cultural norms influence how leader political be- et al. 2019; Landells and Albrecht 2019; Hochwarter et al. havior is perceived. In some cultures, political behavior 2020). There are studies that attempt to cover both the- may be viewed as acceptable and even expected, while in oretically as well as empirically to better understand the others, it may be seen as unethical or detrimental to trust mechanism regarding the relationship between the politi- and cooperation. The findings through multilevel analyzes cal skills and the effectiveness of leaders. Treadway et al. in this study demonstrated that the initial predictions that (2004) argue that the idea of leader political skill is fun- political skill affects both cognitive and emotional confi- damentally multi-layered. This approach is in line with dence in a similar way, was not realized. Although Turkish current requests for leadership theory that more truly por- society exhibits vertical collectivist traits (emphasizing trays the nested character of organizational phenomena. hierarchy). cultural differences and their implications for Additionally, they (Treadway et al. 2004:493) elaborate persuasive appeals (Shavitt et al. 2008), we conjecture that “that political skill is one of the most important compe- these tend to lean towards vertical individualist behaviors, tencies leaders can possess, contributing to effectiveness where people tend to be concerned with improving their in organizations”. In a meticulous recent meta-analysis, it individual status and with distinguishing themselves from is claimed that obtaining relevant political skill appears to others via competition (Shavitt et al. 2011). In addition, be advantageous for leaders to enhance their effectiveness the study reveals that political skills were found to affect by either obtaining or developing it (Ferris et al. 2019). But both cognitive and emotional confidence. As so, the as- interestingly, it is concluded that to yet, no empirical study sumption that individuals in organizations will have high has been undertaken to evaluate theories about the impli- perceptions of trust towards each other was not found de- cations of a leader’s political competence on employees pending on emotional components and/or on events in so- (Treadway et al. 2004). cieties showing collectivist characteristics, the study found Thirdly, we contribute to the argument on the percep- no significant difference between cognitive trust and emo- tion of leadership within groups a multilevel model that tional trust perceptions. The reason for this may be that evaluates within- and between-level variance, helping Turkish society no longer has dominant vertical collectiv- to better understand the collective, dyadic, nested, and ist characteristics as claimed in past studies, in which peo- within-person effects that have been noted by Mahon and ple focus on complying with authorities and on enhancing Greenwald (2018) to be sparse within the literature. The the cohesion and status of their in-groups, even when this study provides that leader prototype perceptions of em- entails sacrificing their own personal goals (Shavitt et al. ployees increased their perceptions of trust in the leader; 2011). Findings show that individuals tend to develop the moreover, the perceptions of trust in the leader had a full perception of trust based on the constructive and develop- mediating effect between the leader prototype perception ing behaviors of leaders (personal empowerment). towards and the leader effectiveness perceptions. The inclusion the employees despite the leader’s political discourse and of leader’s political skills utilizing a multilevel analyzes behavior. The causes of this is conjectured to the lack of added great value to the findings and to the originality of dominant vertical collectivist characteristics in Turkish the research. As so, the findings provide validity of social society and to individual tendency to develop perception identity theory outside of individual western societies (van of trust based on the leader’s actions and their sense of 357 Organizacija, Volume 57 Research Papers Issue 4, November 2024 personal empowerment, despite the leader’s political dis- developing trust. course and behavior. Consequently, cultural characteristics This research is subject to some potential limitations. of Turkish society can be a reference for effective solutions First, the study results may be susceptible to same-source in business management. In collectivistic cultures like Tur- bias because all variables were collected from the study key, group harmony and conformity are prioritized, leader participants, consisting of service sector employees and political behavior that benefits the group may be seen as managers in Ankara through site visit surveys. However, more acceptable, even if it involves strategic maneuvers or the study’s design minimized the potential for this bias giv- power plays. Whereas in individualistic cultures, where in- en that we implemented data collection to multiple organ- dividual autonomy and personal interests are emphasized, izations, two mid-sized and three large sized organizations leader political behavior may be viewed more skeptical- from the surrounding provinces and districts of Ankara. ly, as it may be perceived as self-serving or manipulative. In addition, the surveys were conducted in two separate Particularly for managers and companies running in col- time periods, using time lag and variables from multiple lectivist societies like Turkey, the findings of this study sources can reduce the threat of common source and meth- have important pragmatic consequences. Leaders should od variance (Podsakoff et al. 2003). Second, this research understand that their performance and the confidence they was conducted in Turkey, which allows us to examine inspire from their subordinates depend much on their polit- leadership prototype perception in context characterized ical abilities and conformity with organizational standards. by high context values and its citizens are traditionally These findings allow managers to create training courses accustomed to collectivist, high power distance schemas improving political awareness and promoting a trusting (Hofstede 1980, 2011); however, there are vastly different society. Furthermore, companies can consider cultural as- constraints and experiences in the diverse socio-cultural pects in planning leadership development programs to fit and geo-political contexts of countries classified as high the particular requirements and expectations of their em- context cultures. Generalizing the findings of this study to ployees. other high context cultures would require caution as there are vast differences between them. Future research may 6 Conclusion examine the study model in other cultural contexts. Future studies should investigate, in different cultural settings, the long-term effects of political behaviors and In this paper, empirical evidence dealing with how a leadership prototypes on organizational outputs. Further- leader’s political skills may act as a moderator that might more, looking at how various forms of trust—cognitive decrease the effect of prototypicality’s impact on trust and and affective—as well as leadership effectiveness interact on leadership effectiveness was investigated. The study in various cultural contexts would help one to grasp the employed a scale of trust used to measure leader’s cogni- fundamental processes. Research might also look at how tive and emotional dimensions, accepted to be compatible views and behaviors of leaders are shaped by outside en- with social identity theory. The findings put forth utilizing vironmental elements such as society standards and eco- social identity theory demonstrates how the leading pro- nomic situation. All things considered, this paper offers totypes formed in the cognitive schemes of individuals insightful analysis of how political skills and leadership reflect the values and norms of the organization, which prototypes shape leadership effectiveness and confidence can build trust in the leader, contributing towards positive in the Turkish setting. We have shown using a multilevel aspects of leader political skills rarely observed or studied moderated mediation model that political skills are a ma- within the field. Through the combining of leadership ef- jor moderator influencing trust and effectiveness of leader fectiveness and leadership political behavior, implications prototypicality. These results challenge mostly negative from a multilevel study towards the need to have more opinions and emphasize the good features of political leadership effectiveness and political leadership skills in- conduct in leadership, therefore augmenting the body of vestigations as a resource for leadership development has knowledge already in use. Furthermore, the study empha- been concluded and brought to light. Furthermore, the sizes the need of cultural background in leadership re- use of aggregation and multilevel modeling has allowed search and provides useful consequences for managers and evaluating and studying collective, dyadic, embedded, companies trying to improve the trust in collectivist socie- and within-person effects within the literature. This study ties by means of leadership effectiveness enhancement. As showed leader’s emotional relationship on its own does a concluding thought with no revelation intended, leaders not provide their followers a sense of empowerment and in Turkish organizational life and we conjecture elsewhere development nor does it produce a high level of emotional as well benefit when leaders develop their political skills trust. In this respect, values related to support, develop- aligned to the goals of the organization and the people they ment and guidance of employees by leaders within Turkish lead, in which their followers’ expectations and needs take organizational culture compares to western organizational forefront. norms (Steffens et al. 2014), and appears to have a part in 358 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Literature ber exchange”, European Journal of Work and Organi- zational Psychology, Vol. 29 No. 1, pp. 106-125. Ferris, G.R., and Hochwarter, W.A. (2011), Organizational Ari, G.S. (2003), The Relationship between Trust and Em- politics. In S. Zedeck (Ed.), APA handbook of indus- ployee Empowerment in Businesses: A Study in the trial and organizational psychology, Vol. 3. Maintain- Banking Sector, Unpublished PhD Thesis, Institute of ing, expanding, and contracting the organization (pp. Social Sciences, Hacettepe University, Ankara. 435–459), American Psychological Association. Baron, J.N., and Kenny, D.A. (1986), “The moderator-me- Ferris, G.R., Treadway, D.C., Kolodinsky, R.W., diator variable distinction in social psychological re- Hochwarter, W.A., Kacmar, C.J., Douglas, C., and search: Conceptual, strategic, and statistical consider- Frink, D.D. (2005), “Development and validation of ations”, Journal of Personality and Social Psychology, the political skill inventory”, Journal of Management, Vol. 51 N.6, pp. 1173–1182. Vol. 31 No. 1, pp. 126-152. Barreto, N.B., and Hogg, M.A. (2017), “Evaluation of and Ferris, G.R., Ellen, B.P., McAllister, C.P., and Maher, L.P. support for group prototypical leaders: A meta-analy- (2019), “Reorganizing Organizational Politics Re- sis of twenty years of empirical research”, Social Influ- search: A Review of the Literature and Identification of ence, Vol. 12 No. 1, pp. 41-55. Future Research Directions”, Annual Review of Orga- Barth-Farkas, F., and Vera, A. (2019), “Leader prototyp- nizational Psychology and Organizational Behavior, icality and displayed power in the police: an empiri- Vol. 6, 299-323. cal analysis of the impact on leader endorsement and Ferris, G.R., and Treadway, D.C. (2012), Politics in Orga- trust”, Policing: A Journal of Policy and Practice, Vol. nizations: Theory and Research Considerations. New 13 No. 4, pp. 483-497. York: Routledge/Taylor and Francis. Bliese, P.D. (2000), Within-group agreement, non-inde- Fiske, S.T., and Depret, E. (1996), “Control, interdepen- pendence, and reliability: Implications for data aggre- dence and power: Understanding social cognition in its gation and analysis. In KJ Klein and SWJ Kozlowski social context”, European Review of Social Psycholo- (Eds.), Multilevel theory, research, and methods in gy, Vol. 7 No. 1, pp. 31-61. organizations (pp. 349–381), San Francisco: Jossey- Giessner., SR, van Knippenberg, D., and Sleebos, E. Bass. (2009), “License to fail? How leader group prototypi- Chang, C., Rosen, C.C., and Levy, P.E. (2009), “The Re- cality moderates the effects of leader performance on lationship between Perceptions of Organizational Pol- perceptions of leadership effectiveness”, The Leader- itics and Employee Attitudes, Strain and Behavior: A ship Quarterly, Vol. 20 No. 3, pp. 434–451. Meta Analytic Examination”, The Academy of Man- Glick, W.H. (1985), “Conceptualizing and measuring or- agement Journal, Vol. 52 No. 4, pp. 779-801. ganizational and psychological climate: Pitfalls in Colquitt, J.A., Scott, B.A., and LePine, J.A. (2007), “Trust, multilevel research”, Academy of Management Re- trustworthiness, and trust propensity: a meta-analytic view, Vol. 10 No. 3, pp. 601–616. test of their unique relationships with risk taking and Goodwin, V.L., Whittington, J.L., Murray, B., and Nich- job performance”, Journal of Applied Psychology, Vol. ols, T. (2011), “Moderator or mediator? Examining the 92 No. 4, pp. 909-927. role of trust in the transformational leadership para- de Vries, M.K., and Cheak-Baillargeon, A. (2015), Lead- digm”, Journal of Managerial Issues, Vol. 23 No. 4, ership in organizations, sociology of, in Wright, J.D. pp. 409-425. (ed), International Encyclopedia of the Social and Be- Guillen, L., Mayo, M., and Korotov, K. (2015), “Is lead- havioral Sciences (Second Edition), pp. 664-669. ership a part of me? A leader identity approach to un- Dirks, K.T., and Ferrin, D. L. (2002), “Trust in leadership: derstanding the motivation to lead”, The Leadership Meta-analytic findings and implications for research Quarterly, Vol. 26 No. 5, pp. 802-820. and practice”, Journal of Applied Psychology, Vol. 87 Greenberg, J., Ashton-James, C.E., and Ashkanasy, N.M. No. 4, pp. 611–628. (2007), “Social comparison processes in organiza- Ellen, III B.P., Ferris, G.R., and Buckley, M.R. (2013), tions”, Organizational Behavior and Human Decision “Leader political support: Reconsidering leader politi- Processes, Vol. 102 No. 1, pp. 22–41. cal behavior”, The Leadership Quarterly, Vol. 24 No. Halevy, N., Chou, E.Y., and Galinsky, A.D. (2011), “A 6, pp. 842-857. functional model of hierarchy: Why, how, and when Ellen, III B.P. (2014), “Considering the positive possibili- vertical differentiation enhances group performance”, ties of leader political behavior”, Journal of Organiza- Organizational Psychology Review, Vol. 1 No. 1, pp. tional Behavior, Vol. 35 No. 6, pp. 892-896. 32–53. Fein, E.C., Benea, D., Idzadikhah, Z., and Tziner, A. Hasel, M.C. (2013), “A Question of Context: The Influ- (2020), “The security to lead: a systematic review of ence of Trust on Leadership Effectiveness During Cri- leader and follower attachment styles and leader–mem- sis”, Management, Vol. 3 No. 16, pp. 264-293. 359 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Hochwarter, W.A. (2012), The positive side of organi- ward”, Journal of Business and Psychology, Vol. 38 zational politics. In GR Ferris and DC Treadway No. 2, pp. 239-258. (Eds.), SIOP organizational frontiers series. Poli- Liden, R.C., Wu, J., Cao, A.X., and Wayne, S.J. (2015), tics in organizations: Theory and research consid- The Oxford handbook of leader-member exchange, erations (pp. 27–65), Routledge/Taylor and Francis Oxford University Press, UK. Group. Lord, R., and Hall, R. (2003), Identity, leadership catego- Hochwarter, W.A., Rosen, C.C., Jordan, S.L., Ferris, G.R., rization, and leadership schema. In Van Knippenberg, Ejaz, A., and Maher, L.P. (2020), “Perceptions of orga- D. and Hogg, M.A., editors, Leadership and power: nizational politics research: past, present, and future”, Identity processes in groups and organizations. Lon- Journal of Management, Vol. 46 No. 6, pp. 879-907. don, UK: Sage Publications, pp. 48–64. Hofstede, G. (1980), Culture’s consequences: Internation- Lvina, E., Johns, G., Treadway, D.C., Blickle, G., Liu, al differences in work-related values. Beverly Hills, Y.L., Liu, L., Atay, S., Zettler, I., Solga, J, Noethen, CA: Sage. D., and Ferris, G.R. (2012), “Measure invariance of the Hofstede, G. (2011), “Dimensionalizing cultures: The Political Skill Inventory (PSI), across five cultures”, Hofstede model in context”, Psychology and Culture, Intercultural Management of Cross-cultural Manage- Vol. 2 No. 1, pp. 1-25. ment, Vol. 12 No. 2, pp. 171-191. Hogg, M.A. (2001), “A Social Identity Theory of Leader- Mahon, M., and Greenwald, J.M. (2018), “The influence ship”, Personality and Social Psychology Review, Vol. of leader prototype congruence on leadership self-ef- 5 No. 3, pp. 184-200. ficacy and motivation to lead”, Academy of Manage- Hogg, M.A., Abrams, D., Otten, S., and Hinkle S. (2004), ment Proceedings, Vol. 1, 12099. “The social identity perspective: Intergroup relations, Mayer, R.C., Davis, J.H. and Schoorman, F.D. (1995), self-conceptions, and small groups”, Small Group Re- “An Integrative Model of Organizational Trust”, The search, Vol. 35 No. 3, pp. 246–276. Academy of Management Review, Vol. 20 No. 3, pp. Ibarra, H., Snook, S., and Guillén Ramo, L. (2010), Iden- 709-734. tity-based leader development. In R. Khurana and N. Mayseless, O. (2010), “Attachment and the leader—fol- Nohria (Eds.),, Leadership: Advancing an intellectual lower relationship”, Journal of Social and Personal discipline (pp. 657–678), Boston, MA: Harvard Busi- Relationships, Vol. 27 No. 2, pp. 271-280. ness Press. McAllister, D.J. (1995), “Affect- and Cognition-Based Jacquart, P., and Antonakis, J. (2015), “When does charis- Trust as Foundations for Interpersonal Cooperation in ma matter for top-level leaders? Effect of attributional Organizations”, The Academy of Management Jour- ambiguity”, Academy of Management Journal, Vol. 58 nal, Vol. 38 No. 1, pp. 24-59. No. 4, pp. 1051-1074. Mehmood, Q., Hamstra, M.R.W., and Schreurs, B. (2019), James, D.L., Demaree, R.G., and Wolf, G. (1984), “Es- “Employees’ perceptions of their manager’s authentic timating within-group interrater reliability with and leadership: Considering managers’ political skill and without response bias”, The Journal of Applied Psy- gender”, Personnel Review, Vol. 49 No. 1, pp. 202-214. chology, Vol. 69 No. 1, pp. 85–98. Morgeson, F.P., DeRue, D.S., and Karam, E.P. (2010), Kapoutsis, I., and Thanos, I. (2016), “Politics in organiza- “Leadership in teams: A functional approach to under- tions: positive and negative aspects of political behav- standing leadership structures and processes”, Journal ior”, European Management Journal, Vol. 34 No. 3, of Management, Vol. 36 No. 1, pp. 5–39. pp. 310-312. Muthén, B., and Asparouhov, T. (2009), “Multilevel re- Kastanakis, M., and Voyer, B.G. (2014), “The effect of cul- gression mixture analysis”, Journal of the Royal Sta- ture on perception and cognition: a conceptual frame- tistical Society: Series A, Vol. 172 No. 3, pp. 639–657. work”, Journal of Business Research, Vol. 67 No. 4, Ömür, M. (2018), The Role of Leader›s In-Group pp. 425-433. Prototypicality and Voter Characteristics in Klein, K.J., and Kozlowski, S.W.J. (2000), Multilevel the- Evaluations of Political Party Leaders, Unpublished ory, research, and methods in organizations: Founda- PhD Thesis, Institute of Social Sciences, Ankara tions, extensions, and new directions. San Francisco: University, Ankara. Jossey-Bass. Özdemir, M., and Gören, S.Ç. (2015), “Politik beceri en- Landells, E.M., and Albrecht, S.L. (2019), “Perceived vanterinin eğitim örgütlerinde geçerlik ve güvenirlik organizational politics, engagement, and stress: The çalışmaları”, Kuram ve Uygulamada Eğitim Yönetimi, mediating influence of meaningful work”, Frontiers in Vol. 21 No. 4, pp. 521-536. Psychology, Vol. 10 No. 10, pp. 1612. Platow, M.J., Haslam, S.A., Foddy, M., and Grace, D.M. LeBreton, J.M., Moeller, A.N., and Wittmer, J.L. (2023), (2003), Leadership as the outcome of self-categoriza- “Data Aggregation in Multilevel Research: Best Prac- tion processes. Leadership and power: identity pro- tice Recommendations and Tools for Moving For- cesses in groups and organizations. London, United 360 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Kingdom: Sage Publications. Pp. 34-47. shared identity impacts group members’ effort and Podsakoff, P.M., MacKenzie, S.B., and Bommer, W.H. performance: Evidence from an exercise task”, PLoS (1996), “Transformational leader behaviors and sub- ONE, Vol. 14 No. 7, pp. e0218984. stitutes for leadership as determinants of employee Sun, L.Y., Zhang, Z., Qi, J., and Chen, Z.X. (2012), “Em- satisfaction, commitment, trust, and organizational powerment and creativity: A cross-level investigation”, citizen”, Journal of Management, Vol. 22 No. 2, pp. The Leadership Quarterly, Vol. 23 No. 1, pp. 55-65. 259–298. Tajfel, H. (1972), Social categorization (English transla- Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y., and Podsa- tion of “La categorisation sociale”). In Moscovici, S., koff, N.P. (2003), “Common method biases in behav- editors. Introduction d la psychologie sociale, vol. 1. ioral research: a critical review of the literature and Paris: Larousse, pp. 272-302. recommended remedies”, Journal of Applied Psychol- Treadway, D.C., Hochwarter, W.A., Ferris, G.R., Kacmar, ogy, Vol. 88 No. 5, pp. 879. C.J., Douglas, C., Ammeter, A.P., and Buckley, M.R. Preacher, K.J., Zyphur, M.J., and Zhang, Z. (2010), “A (2004), “Leader political skill and employee reac- general multilevel SEM framework for assessing mul- tions”, The Leadership Quarterly, Vol. 15 No. 4, pp. tilevel mediation”, Psychological Methods, Vol. 15 493-513. No. 3, pp. 209–233. van Dick, R., and Kerschreiter, R. (2016), “The Social Ridgeway, C.L. (2001), Social status and group structure. Identity Approach to Effective Leadership: An Over- In M.A. Hogg and R.S. Tindale (Eds.),, Blackwell view and Some Ideas on Cross-Cultural Generalizabil- handbook of social psychology: Group processes (pp. ity”, Frontiers of Business Research in China, Vol. 10 352-375), Oxford, England: Blackwell. No. 3, pp. 363–384. Schneider, B., White, S.S., and Paul, M.C. (1998), “Link- van Knippenberg, B., and van Knippenberg, D. (2005), ing service climate and customer perceptions of ser- “Leader self-sacrifice and leadership effectiveness: the vice quality: Test of a causal model”, Journal of Ap- moderator role of leadership prototypicality”, Journal plied Psychology, Vol. 83 No. 2, pp. 150–163. of Applied Psychology, Vol. 90 No. 1, pp. 25 – 37. Schwartz, S.H. (2006), “A theory of cultural value orien- Van Knippenberg, D. (2011), “Embodying who we are: tations: Explications and applications”, Comparative Leader group prototypicality and leadership effective- Sociology Vol. 5 No. 2/3, pp. 137–182. ness”, The Leadership Quarterly, Vol. 22 No. 6, pp. Shavitt, S., Lee, A., and Johnson, T.P. (2008), Cross-cul- 1078–1091. tural consumer psychology. In Haugtvedt, C., Herr, P., Zibenberg, A. (2017), “Perceptions of Organizational Pol- Kardes, F., editors. Handbook of Consumer Psycholo- itics: A Cross-cultural Perspective”, Global Business gy. Mahwah, N.J: Erlbaum, pp. 1103–1131. Review, Vol. 18 No, 4, pp. 1-12. Shavitt, S., Johnson, T.P., and Zhang, J. (2011), “Horizon- tal and Vertical Cultural Differences in the Content of Advertising Appeals”, Journal of International Con- Aykut Arslan is a Professor of Business and Strategic sumer Marketing, Vol. 23 No. 3-4, pp. 297–310. Management and the Dean of Economics and Steffens, N.K., Haslam, S.A., Reicher, S.D., Platow, M.J., Administrative Sciences Faculty at Piri Reis University Fransen, K., Yang, J., Ryan, M.K., Jetten, J., Peters, in Istanbul. He has published in Journal of Business K., and Boen, F. (2014), “Leadership as social identity Ethics, Journal of Workplace Behavioral Health, management: Introducing the Identity Leadership In- Procedia-Social and Behavioral Sciences, Culture and ventory (ILI) to assess and validate a four-dimensional Organization and Current Psychology among others. model”, The Leadership Quarterly, Volume 25 No. 5, pp. 1001-1024. Steffens, N.K., Munt, K.A., van Knippenberg, D., Platow, Serdar Yener is an Associate Professor in the Faculty M.J., and Haslam, S.A. (2021), “Advancing the social of Economics and Administrative Sciences at Sinop identity theory of leadership: A meta-analytic review University. He has published in International Journal of Management Practice, Sustainability, International of leader group prototypicality”, Organizational Psy- Journal of Social and Educational Sciences, Journal of chology Review, Vol. 11 No. 1, pp. 35-72. Nursing Management and Current Psychology among Steffens, N.K., Peters, K., Haslam, S.A., and Platow, M.J. others. (2018), “One of us… and us… and us: Evidence that leaders’ multiple identity prototypicality (LMIP), is related to their perceived effectiveness”, Comprehen- Fatma Korkmaz is an Associate Professor in the sive Results in Social Psychology, Vol. 3 No. 2, pp. Department of Business Administration at Yozgat Bozok 175-199. University. She has published in Journal of Business Stevens, M., Rees, T., Steffens, N.K., Haslam, S.A., Cof- Studies, International Journal of Social Studies, and fee, P., and Polman, R. (2019), “Leaders’ creation of 361 Organizacija, Volume 57 Research Papers Issue 4, November 2024 International Journal of Business and Management Culture and Organization, Human Resource Studies among others. Development International, International Journal of Gender & Entrepreneurship, Journal of Enterprising Culture, Innovation: Management, Policy and Practice, Murat S. Erogul is an Associate Professor in the Current Psychology, and Journal of Developmental Department of Management at Adelphi University. Entrepreneurship among others. He has published in Gender, Work and Organization, Večnivojska preiskava dojemanja prototipa vodenja: politično vedenje v povezavi z učinkovitostjo in zau- panjem Ozadje/namen: Empirični dokazi kažejo, da lahko voditeljeve politične sposobnosti delujejo kot moderator, ki zmanjšuje vpliv prototipičnosti na zaupanje in učinkovitost vodenja. Študija preučuje vodilne politične veščine kot drugostopenjske regulativne spremenljivke v povezavi z učinkovitostjo in zaupanjem vodje, z namenom testiranja mednivojskega moderiranega modela mediacije znotraj tradicionalno kolektivistične kulture, kot je Turčija. Cilj je kontekstualizirati in razširiti razumevanje dojemanja prototipa vodenja. Metode: Anketiranih je bilo 442 zaposlenih v storitvenem sektorju in 28 vodilnih delavcev. Izvedeni sta bili dve razi- skavi v dveh ločenih časovnih obdobjih. Za ovrednotenje hipotez je bila uporabljena večnivojska analiza poti. Rezultati: Ugotovitve kažejo, da obstaja močan in pomemben vpliv prototipičnosti vodje na učinkovitost vodenja ter neposreden vpliv na zaupanje. Prav tako se izkaže, da so vodje, ki jim podrejeni zaupajo, v svojih vodstvenih vlogah bolj učinkoviti. Nižja raven zaupanja podrejenih je negativno povezana z učinkovitostjo vodje. Zaključek: Če vodje svojim sledilcem ne morejo zagotoviti kontekstualiziranega občutka opolnomočenja in razvoja, to pomeni, da ne zagotavljajo visoke ravni čustvenega zaupanja. Zato obstaja večja potreba po večnivojskih konte- kstualiziranih študijah, ki upoštevajo kolektivne, dvostranske in vgrajene izkušnje znotraj skupin. Ključne besede: Politično vedenje voditelja, Politične veščine, Učinkovitost vodje, Zaupanje, Teorija družbene iden- titete, Turčija 362 Organizacija, Volume 57 Research Papers Issue 4, November 2024 DOI: 10.2478/orga-2024-0026 Government Effectiveness in the Petroleum Sector: Two-step Analysis Combining Linear Regression and Artificial Neural Networks Barbara DORIĆ1, Dinko PRIMORAC2, Mirjana PEJIĆ BACH3 1 University of Ljubljana, Faculty of Economics and Business, Ljubljana, Slovenia, barbaradoric@yahoo.com 2 University North, Croatia, Koprivnica, Croatia, dprimorac@unin.hr 3 University of Zagreb, Faculty of Economics & Business, Zagreb, Croatia, mpejic@net.efzg.hr Background and Purpose: To encourage petroleum industry development, a country needs to set up a regulatory framework that standardizes investment conditions. The objective of the research was to investigate the determi- nants of government effectiveness in the petroleum sector. Design/Methodology/Approach: Multiple regression analysis was conducted to investigate if government effec- tiveness in the petroleum sector is influenced by the country’s political stability, regulatory quality, the intensity of petroleum exploration and production activities, government take, and type of contract used. Artificial neural network analysis was additionally conducted to identify the importance of independent variables. Results: Political stability, regulatory quality, government take attractiveness, and the intensity of petroleum ac- tivities positively influence government effectiveness. A more attractive government take enhances effectiveness, while the type of contract for awarding petroleum rights did not significantly impact effectiveness. Artificial neural network analysis revealed that the most important variables were regulatory quality and political stability. Conclusion: The research concluded that political stability, regulatory quality, and the intensity of petroleum activ- ities are key factors in enhancing government effectiveness in the petroleum sector. These findings have practical implications, as they emphasize the importance of stable and well-regulated environments for achieving higher government effectiveness in the petroleum industry. This equips policymakers and industry professionals with ac- tionable insights for improving the sector’s performance. Keywords: Energy policy, Government effectiveness, Petroleum sector performance, Petroleum resources manage- ment, Industry development 1 Introduction lished a legal practice regarding the rights of oil leases and the management of revenue from its production. The 1889 As petroleum production expanded in the United Pennsylvania Supreme Court decision equated the produc- States, disputes arose regarding land ownership and shares tion of oil and gas to that of other minerals, concluding in profit (Hammerson, 2011). American state courts estab- that land ownership does not necessarily entail ownership 1 Received: 20th December 2023; Accepted: 22nd August 2024 363 Organizacija, Volume 57 Research Papers Issue 4, November 2024 of minerals (Hammerson, 2011). Texas applied the offset ing one of the most important primary resources, and the rule for neighboring wells and the concept of ownership in use of energy became important in contemporary industri- place, which defined the ownership principles in petroleum al infrastructure. This increased value influenced the codi- production as either freehold ownership of the land, which fication and regulation of relationships among participants included the right to minerals, or partial ownership, which in petroleum exploration and production. did not include the right to minerals (Thurman, 2022). In countries with a free market economy, petroleum Capital investments in petroleum exploration and produc- companies conducting petroleum exploration and produc- tion and the return on investment in this activity could be tion activities were state-owned, forming part of a planned compared to the riskiest investments in speculative trends and targeted economy. Following the disintegration of a on the capital market (Simkins & Simkins, 2013). Despite non-market and planned socio-economic system, free cap- this, the possibility of exceptionally high profits in the case ital ownership has become a global universal principle of of a positive petroleum discovery motivated oil companies relationships. Ownership and contractual relationships in to take such risks. the area of exploration and production of valuable natural The United States legal system is based on the prin- resources with high capital intensity and value, such as pe- ciples of Anglo-Saxon law and precedents, and court troleum, have become a matter of special attention for all decisions have also established the legal practice for re- government instances. lationships among participants in petroleum exploration Laws and legal regulations regarding petroleum pro- and production activities. The starting point is the freehold duction were once part of mining legislation. However, ownership category, which, along with the land ownership, since petroleum is present in the Earth’s crust in varying entails the right to minerals, i.e., oil and gas. In contrast, physical and geological forms, exploration is performed European countries implemented different forms of feudal using a range of technical means, and the production tech- and royal limitations regarding mining rights. The owner- nology differs from that in the production of solid mineral ship of petroleum in European countries is considered a raw materials. Petroleum legislation sets out conditions for public good and is regulated by provisions governing state investments in petroleum exploration and production, le- ownership (Thurman, 2022). gal prerequisites for development, and competitive terms The principal dissimilarity between the petroleum regarding petroleum exploration and production (John- exploration and production business in the United States ston, 1994). It places significant emphasis on optimization and the rest of the world stems from the definition of min- during mineral raw material management processes while eral ownership (Seba, 2008). In countries applying An- primarily protecting national interests and providing pe- glo-Saxon law, oil leases are based on freehold ownership, troleum companies (investors) with security and stability which includes the right to minerals, whereby the lease as they carry out their investments and business activities includes compensation for the land and part of the value (Green & Smith, 2023; Thurman, 2022). of the produced oil and gas. In most countries worldwide, The regulatory framework in every country is based on where state ownership of minerals prevails, oil compa- the nation’s constitution, which grants taxing and legisla- nies acquire the right to minerals from the government. At tive authority that governs petroleum legislation and out- the same time, the lease for the use of the land is agreed lines authority boundary conditions for relationships with upon with landowners based on local laws and regulations foreign companies. The function of government is to pro- (Simkins & Simkins, 2013). The relationships between the vide an adequate regulatory infrastructure for companies company acquiring the rights to minerals and the previous, to work economically productive units and ensure they i.e., original, owner (freehold owner in the Anglo-Saxon do not swindle the public, exploit workers, pollute their law or the state in continental law) are governed by a con- surroundings, prosecute unethically, or engage in morally tract defining the terms and compensation for rights to pe- or socially reprehensible practices (Parra, 2004). Hence, troleum. This is a specific compensation, income, or yield establishing a regulatory framework that standardizes in- obtained by the landowner (state) and represents a cost for vestment conditions is an essential step in promoting the the petroleum lessee, different than all other taxes or ex- growth of the petroleum sector. penses. This yield is known as a royalty, i.e., the fee for The study aims to examine the factors that influence recovered quantities of petroleum. In the United States, it government efficiency in the petroleum industry. This traditionally amounts to 1/8 (12.5%) of the market value of study employs multiple regression analysis (MLA) to ex- the produced petroleum (Johnston, 1994). amine the potential impact of political stability, regulatory In the 20th century, contractual relationships in pe- quality, the intensity of petroleum exploration and produc- troleum exploration and production developed due to the tion operations, government take, and type of contract on rise in petroleum production and exploration and rising oil government performance in the petroleum industry. Fur- prices. Oil and gas became essential primary sources of thermore, a study of artificial neural networks (ANN) was energy, accounting for over two-thirds of primary energy performed to determine the significance of independent consumption. The expansion of transport led to oil becom- factors. 364 Organizacija, Volume 57 Research Papers Issue 4, November 2024 2 Theoretical background Smith, 2023). The fiscal regime, if balanced and regulated proper- Petroleum legislation sets out conditions for invest- ly, can attract significant investments in exploration and ments, legal prerequisites for development, and com- production activities and create wealth for the nation. The petitive terms regarding petroleum exploration and pro- higher the government take, the greater the probability of duction. It emphasizes optimization during mineral raw creating wealth for the nation. To determine the attractive- material management processes and provides security and ness of the fiscal regime, the government take is combined stability for petroleum companies as they carry out their with other measures of profitability, including fiscal sys- investments and business activities. tem flexibility, revenue risk, and fiscal stability (Johnston, The petroleum regulatory framework in a country is 2003). The most common fiscal regime terms used world- based on its constitution, which grants taxing and legis- wide are bonuses, fees, state participation, royalty, produc- lative authority for petroleum legislation and outlines au- tion sharing, cost recovery, and taxes (Simkins & Simkins, thority boundary conditions for relationships with foreign 2013). The terms of the fiscal regime differ among coun- companies. The constitution also includes specific petro- tries, and not all are included within one particular regime. leum legislation that authorizes the national oil compa- Production sharing is a fiscal regime that allows reve- ny or responsible ministry to negotiate certain aspects of nue from petroleum production to be shared between the agreements between the state and foreign companies. Tax domicile country and the oil company, allowing the com- liabilities are usually included in the agreement signed be- pany to recover costs and make a return on investment tween the parties (government and petroleum company) (Johnston, 2003). The three main elements of production and regulated by separate laws. sharing are cost recovery, excess cost recovery, and profit Governments provide an adequate regulatory infra- share. Taxes are common to both fiscal regime systems, structure for companies to work economically productive including corporate income tax, additional profit taxes de- units and ensure they do not swindle the public, exploit fined only for petroleum operation companies operating in workers, pollute the environment, prosecute unethically, or the domicile country, and dividend withholding taxes. defraud shareholders. Changes in the economic environ- The ideal fiscal regime should ensure a stable business ment and the increasing interest of foreign investors have environment, minimize sovereign risk, discourage undue indicated the need to regulate petroleum exploration and speculation, provide the potential for a fair return, balance production in a manner defined and accepted within global risk and reward, avoid complexity, limit administrative practice. Petroleum legislation needs to create conditions burden, allow flexibility, and promote healthy competition for large investments, determine the legal prerequisites for and market efficiency (Johnston, 2003). The most com- energy development, and protect national interests in the mon petroleum industry-recognized fiscal regimes fall petroleum sector. broadly into two categories: the concession system, which Petroleum lease contracts are more similar to financial includes special fees and taxes payable in money to the contracts than typical land concessions or mining conces- country where it is operating, and the contractual (produc- sions due to the uncertainty of petroleum prices and the in- tion sharing) system, which includes production sharing creased strategic role of petroleum. Both parties have indi- arrangements where petroleum is usually shared in kind vidual interests, with the oil company minimizing risk and between oil company and domicile country (Simkins & the government increasing its share in profit distribution. Simkins, 2013). This results in direct increases in fiscal revenue through In the concessionary system, oil companies have the royalties, taxes, and indirect contributions. right to perform petroleum exploration and production at The fiscal regime, or petroleum taxation model, is a their own cost, assuming the overall risk of discovery and financial structure that oil companies must pay to countries production risks. The royalty goes directly to the country for petroleum exploration and production activities. It is as one part of the country’s petroleum profit, and all tax- often represented as a government take versus an oil com- es are payable on profit before income tax (Seba, 2008). pany take, with the government taking the percentage of Figure 1 shows the typical revenue distribution under the profit that goes directly to the state budget and the oil com- concession system and illustrates the hierarchy of royal- pany taking the percentage remaining with the company ties, deductions, and examples of possible taxation layers. (Johnston, 2003). There are 145 countries worldwide with Of the total revenues collected from petroleum produced, specific fiscal and contractual terms for engaging with oil the royalty (percentage of total petroleum value) goes di- companies for petroleum exploration and production oper- rectly to the country as one part of the country’s petroleum ations (IHS Energy, 2016). These regimes can be divided profit (state budget revenue). Before-tax calculations, roy- into two main categories: the concessionary system (based alties are deducted together with all capital and operating on royalty and tax payments) and the production-sharing expenditures (CAPEX and OPEX) from total revenues to system (based on petroleum production sharing) (Green & give the oil company profit before income tax. All taxes (income tax, petroleum special tax, and any other taxes) 365 Organizacija, Volume 57 Research Papers Issue 4, November 2024 are payable on profit before income tax. Income tax goes petroleum, and this is the oil company’s revenue. What re- to the country as the second part of the country’s petroleum mains after cost recovery is the profit share. Profit share is profit (state budget revenue). The remainder after taxes is then split between the country and oil company based on the oil company’s petroleum profit. the contracted percentages. Profit share represents one part Production sharing contracts (PSC) are a newcomer of the country’s petroleum profit (state budget revenue), to the petroleum industry, starting in 1966 in Indonesia while for the oil company, it is one part of the oil com- (Markus, 2014). These contracts involve a contractual pany revenue. From oil company revenue, comprised of relationship between the state and the oil company, with cost recovery and the company’s part of the profit share, all the state owning petroleum rights and the oil company en- capital and operating expenditures (CAPEX and OPEX) suring the execution, technical, and financial realization are deducted. What remains after deductions is subject to of petroleum exploration, development, and production taxation. Income tax is paid to the country, representing (Seba, 2008). The aim is to maximize income and initi- the second part of the country’s petroleum profit (state ate economic activities connected to petroleum explora- budget revenue), while the remainder after taxes is the oil tion and production. Production is split between the host company’s petroleum profit. government and the contractor, with the government Fiscal regimes are often categorized as hybrids, com- maintaining ownership of the produced petroleum. Stabi- bining elements of both classifications, such as royalties lization clauses are essential to ensure the preservation of and taxes. These hybrid systems aim to ensure petroleum the tax system and fiscal proportions throughout the con- profit from the start of production, with most countries tract. Figure 2 shows typical revenue distribution under the using the concessionary system (royalty/tax contract) and production-sharing system and illustrates the calculation production sharing system (production sharing contract). of revenues and costs that would be experienced in a full As shown in Table 1, most countries worldwide use these cycle. From total revenues, collected from the petroleum two systems: concessionary system (royalty/tax contract) produced, cost recovery is deducted first. Cost recovery and production sharing system (production sharing con- includes all capital and operating expenditures (CAPEX tract). and OPEX) borne by the oil company in producing the Source: Doric B. (2017) Figure 1: Concession system - Typical cash flow diagram 366 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Source: Doric B. (2017) Figure 2: Production sharing system - Typical cash flow diagram Table 1: Types of contracts (agreements) used worldwide Type of contract Number of countries Royalty/Tax 78 Production Sharing Agreement 52 Mixed / Various 15 Source: IHS Energy (2016a) Contract terms for oil exploration and production try-standardized and consists of several steps. contracts include fiscal terms, work commitments, insur- Concessionary and production-sharing systems have ance, and local content. Work commitment is crucial for advantages and disadvantages, but the choice of system the host country to ensure the oil company commits to as may not be as critical from an economic, accounting, and much work as possible, enabling quicker development of financial perspective (Johnston, 2003). The fundamental potential new production and revenue generation. Insur- difference between the two systems lies in the ownership ance is essential to ensure high-quality work commitments of the petroleum produced. and cover potential losses. Contracts often require the oil To encourage petroleum industry development, a company to buy goods and services locally to boost local country needs to set up a regulatory framework that will industry development and employ local labor. The recog- standardize the conditions for investments in petroleum nition of exploration and production costs is sensitive, as exploration and production and lay down the legislative the investment of funds and risk lies with the oil company. prerequisites for the development and competitive condi- The country must establish a fiscal regime that maximizes tions in this activity. The function of government is crucial revenues and provides investors with incentives to explore in organizing and managing the petroleum sector since an and develop petroleum efficiently. The license round pro- adequate regulatory framework will ensure that explora- cedure for petroleum exploration and production is indus- tion and production activities are conducted in a way that 367 Organizacija, Volume 57 Research Papers Issue 4, November 2024 will create wealth for the nation, protect the environment, indicates that government take attractiveness should also and enable companies to work in a stable and competi- influence government effectiveness. tive environment (Parra, 2004). Many authors (Falola & The theory outlined in this chapter suggests that nei- Genova, 2005; Shaffer, 2011; Thurber et al., 2011; Holden, ther the type of fiscal regime system nor the correspond- 2013; Kemal, 2016) have argued that political stability is ing type of contract is better or worse. From the economic an important prerequisite for petroleum industry develop- perspective, the same objectives can be achieved through ment and enables economic growth and generates wealth both concessionary and production-sharing contracts. for the nation. Most of the theories to date have been based Therefore, it could be concluded that the type of contract on historical information and statistical observations on has no influence on government effectiveness in the petro- the development of the petroleum industry. Thurber et leum sector. al. (2011) went a step further in their empirical research, indicating that some countries implementing the Norwe- gian Model failed due to a lack of institutional quality and 3 Hypothesis development political stability, which influenced government effective- ness in the petroleum sector. This argument was further According to the theoretical findings outlined in the elaborated by Kemal (2016), who stated that the economic previous chapter, the objective of this empirical study is impact due to changes in petroleum governance might de- to examine the influence of political stability, regulatory pend on political conditions. The literature has indicated quality, the intensity of petroleum exploration and produc- that political stability is an essential factor that directly tion activities, government take (fiscal regime) attractive- influences government effectiveness and thus shows the ness, and the type of contract on government effectiveness. ability of the government to create a stable environment Based on the theory discussed, the following hypotheses when it comes to investments in the petroleum sector. were developed. The theory discussed demonstrated that petroleum leg- The government effectiveness variable represents the islation is the main factor in regulating the complex rela- quality of public service, the quality of civil service and tionship between governments and oil companies in petro- its degree of independence from political pressures, the leum exploration and production activities. It lays down quality of policy formulation and implementation, and the conditions for investments, the legal prerequisites for the credibility of the government’s commitment to such development, and competitive terms and conditions. As policies (World Bank, 2016). Many authors focused on political stability directly dictates the government’s ability government effectiveness when analyzing the petroleum to implement adequate policies and regulations for indus- governance model among various oil-producing countries try development, regulatory quality is another important together with oil-sector performance. Brunnschweiler and factor that can also impact government effectiveness. Bulte (2008) showed that there is a significant difference Exploration activities are crucial for discoveries and in government effectiveness in various oil-producing petroleum production. The more an oil company invests countries. Their empirical research suggested that better in exploration activities, the greater the probability of new government effectiveness led to less resource dependence petroleum discoveries and developing new petroleum pro- and higher direct investments, which in turn positively af- duction. Petroleum exploration intensity depends on geo- fects GDP. Like many other authors, for the government logical probability, which also depends on the investor’s effectiveness variable, they used World Bank data, argu- (oil company) investment in exploration and production ing that these data have the advantage of extensive cov- activities and the owner’s (government) ability to attract erage and objectiveness due to a large survey base, mak- investments and enable exploration and production activi- ing them particularly attractive for econometric analysis ties. Therefore, the intensity of petroleum exploration and (Brunnschweiler & Bulte, 2008). Kaufman et al. (2004) production activities is another factor that may influence argued that the key advantage of the World Bank World- government effectiveness. wide Governance Indicators (WGI) is that despite the mar- Moreover, suppose the relationship between the prom- gins of error, these indicators are sufficiently informative ising petroleum potential and the requirements set in the that many cross-country comparisons result in statistically fiscal regime is unfavorable to oil companies in advance. significant differences in estimated governance. The WGI In that case, they will not proceed with the business and measures six dimensions of governance, which are gov- investments. Thus, governments must design an optimal ernment effectiveness, political stability, regulatory quali- fiscal regime to ensure a favorable balance of mutual rela- ty, rule of law, control of corruption, and voice of account- tions. The government take (share of the petroleum profit) ability. Government effectiveness data published by the is used as a measure to compare the fiscal regimes of dif- World Bank has also been used by other authors (Heller ferent countries in terms of petroleum profit going directly & Marcel, 2012; Thurber et al., 2011), aiming to compare into the state budget and the fiscal regime attractiveness government effectiveness in the petroleum sector among for petroleum sector investments. Accordingly, the theory various oil-producing countries. 368 Organizacija, Volume 57 Research Papers Issue 4, November 2024 The political stability variable represents the level of applied by the country in terms of economic stability and political stability and perceptions of the likelihood of po- balanced share of profits between the country and the oil litical instability and politically motivated violence (World company for 2015. Many authors (Johnston, 2000 and Bank, 2016). Many authors (Falola & Genova, 2005; 2003; Seba, 2008; Thurber et al., 2011; Holden, 2013; Ke- Shaffer, 2011; Thurber et al., 2011; Holden, 2013; Kemal, mal, 2016) have used government take in their empirical 2016) used political stability in their empirical research research and showed that government take attractiveness and showed that the level of political stability would pos- will positively influence government effectiveness in the itively influence petroleum sector governance and conse- petroleum sector. This argumentation leads to the fourth quently government effectiveness in the petroleum sector. hypothesis: The data used in the model was pulled from WGI data pub- • H4: Government take attractiveness positively lished by the World Bank in 2016. Based on this, the first influences government effectiveness in the petro- hypothesis was developed as follows: leum sector. • H1: Political stability positively influences gov- The type of contract represents the contract used ernment effectiveness in the petroleum sector. among selected sample countries when concluding deals The regulatory quality variable represents the per- with oil companies, i.e., the production sharing contract or ception of the ability of the government to formulate and concession (royalty and tax) contract. Some authors, such implement comprehensive policies and regulations that as energy economist Daniel Johnston, have argued that the permit and promote private sector development (World type of contract does not influence the ability of the coun- Bank, 2016). As elaborated in the previous chapter, pe- try to maximize petroleum profit and its effectiveness in troleum legislation is very specific, complex, and indus- the petroleum sector since both types can achieve the same try standardized. Moreover, the ability of the country to objectives. The data used in the model were pulled from implement industry-standardized petroleum legislation is the Petroleum Economics and Policy Solutions (PEPS) very important in terms of the development of the petrole- data published by IHS Energy in 2016. The above indi- um sector. Thus, regulatory quality is another variable that, cates the fifth hypothesis: in addition to political stability, should strongly influence • H5: The type of contract used for awarding petro- government effectiveness. Some of the previous research leum rights does not significantly influence gov- related to the petroleum governance model and respective- ernment effectiveness in the petroleum sector. ly government effectiveness in the petroleum sector used The following chapter outlines the methodology, in- the regulatory quality indicator (Heller & Marcel, 2012; cluding data and analytical approaches, including MLA Thurber et al., 2011). The second hypothesis is developed and ANN. as follows: • H2: Regulatory quality positively influences gov- ernment effectiveness in the petroleum sector. 4 Methodology The intensity of petroleum exploration and production activities variable represents the ranking of the selected 4.1 Data sample countries in terms of the intensity of petroleum ex- ploration and production activities in the five years from After the variables to be used in the model were identi- 2010 to 2016. The ability of the country to attract invest- fied, descriptive statistics were applied, and the variables’ ments in petroleum exploration and production activities descriptions were presented in Table 2. shows the country’s (government) effectiveness in maxi- Government effectiveness data was pulled from WGI mizing its potential and revenues from petroleum (John- data published by the World Bank (2016). The data repre- ston, 1994). This ability can be measured in the intensity of sents an estimation of government effectiveness in each petroleum exploration and production activities. To rank selected country for 2015. Countries were evaluated in the oil-sector performance and measure effectiveness, Thurber range from -2.5 to 2.5, where -2.5 indicates weak effec- et al. (2011) evaluated the ability of the country to develop tiveness and 2.5 indicates strong effectiveness. In order to and produce petroleum, which can only be done through avoid negative values, the range was adjusted by 2.5, to a investments in petroleum operations and showed that this range from 0 to 5, where 0 means weak and 5 means strong ability positively influences petroleum sector effective- effectiveness. ness. The above indicates the third hypothesis: In 2015, countries were evaluated for political stability, • H3: The intensity of petroleum exploration and ranging from -2.5 to 2.5, where -2.5 means weak stability production activities positively influences govern- and 2.5 means strong stability. These ranges were adjusted ment effectiveness in the petroleum sector. by 2.5 to a range of 0 to 5, where 0 means weak, and 5 The government takes the attractiveness variable, means strong stability. which represents the ranking among selected sample Regulatory quality data were also pulled from the WGI countries related to the attractiveness of the fiscal regime data published by the World Bank (2016). The data repre- 369 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 2: Description of the variables Variable Code Measurement Mean St. Dev Government Effectiveness GE 0-5 (0-weak, 5-strong) 2.41 0.956 Political stability PS 0-5 (0-weak, 5-strong) 2.21 0.968 Regulatory quality RQ 0-5 (0-weak, 5-strong) 2.41 0.966 Intensity of exploration and production activities EPI 1-10 (1-low, 10-high) 3.19 2.739 Government take attractiveness GTA 1-10 (1-low, 10-high) 5.98 1.504 Type of contract TC 0-production sharing contract, 0.40 0.492 1-concession (royalty and tax based) contract Source: Authors’ work sents an estimation of regulatory quality in 2015 in each stepwise regression is most commonly used. Stepwise re- selected country. Countries are evaluated in the range from gression can remove any variable whose partial F-value -2.5 to 2.5, where -2.5 means weak quality and 2.5 means indicates that this variable does not contribute, given the strong quality. The range was adjusted by 2.5 to avoid neg- present set of independent variables in the model (Kvanli ative values, and thus, a range from 0 to 5 was used, where et al., 2003). 0 means weak, and 5 means strong quality. In the MLA, many problems may occur due to a large The intensity of petroleum exploration and production number of variables. The purpose of model diagnostics is activities was pulled from the Petroleum Economics and to detect possible weaknesses of the model and, if neces- Policy Solutions data published by IHS Energy (2016). sary, to transform it. Typically, four problems (multicol- Countries were evaluated on a scale of 1 to 10, where 1 linearity, heteroscedasticity, autocorrelation of error terms, means the lowest and 10 means the highest intensity of and the normality of error terms) need to be analyzed in petroleum exploration and production activities. order to prove the validity of the model (Šošić, 2004). If The same source was used to extract the data to meas- any of the four problems are detected, the basic model as- ure government take attractiveness in 2015. Countries sumptions are not satisfied, and the validity of the model were evaluated on a scale of 1 to 10, where 1 means low- is questionable. est, and 10 means highest value. To test the set hypothesis, it was necessary to examine The type of contract was measured as a dummy var- the statistical dependence among variables, which is possi- iable. In the model, countries with production-sharing ble using the MLA. The MLA shows the statistical depend- contracts were denoted with the number 0, while countries ence of one numerical variable (dependent variable) to two with a concession (royalty and tax) contract were denoted or more numerical variables (independent variables). To with the number 1. examine the influence of selected variables on government effectiveness, an MLA was used on a sample of 130 coun- 4.2 Analysis tries worldwide. The dependent variable in the defined model is government effectiveness (GE), and the five in- dependent variables are political stability (PS), regulatory To examine the influence of the defined independent quality (RQ), the intensity of petroleum exploration and variables on the dependent variable, a full MLA is run, in- production activities (EPI), government take attractiveness cluding all independent variables that are considered pre- (GTA), and type of contract (TC). Data were statistically dictors of dependent variables. Since one of the variables analyzed using the programs SPSS 21 and EViews 7. appeared insignificant due to a low t ratio, that variable The correlation matrix is used to check multicollineari- was dropped, and the reduced regression model was rerun ty. The correlation matrix shows the correlation coefficients (Azcel & Sounderpandian, 2009). One of several stepwise between the variables in the model. A serious multicollin- selection procedures is used. These techniques either select earity problem exists if the Pearson coefficient between or eliminate variables, one at a time, in an effort to exclude the variables is 0.9 or greater (Belsey et al., 2004). Other those variables that either have no predictive ability or are multicollinearity problem indicators are variance inflation highly correlated with other predictor variables (Kvanli et factor (VIF) and tolerance indicator (TOL), where VIF>10 al., 2003). Stepwise procedures consist of forward regres- or TOL<0.1 (Hair et al., 1995; Tabachnick & Fidell, 2001; sion, backward regression, and stepwise regression, where Kvanli et al., 2003; O’Brien, 2007). Some authors have 370 Organizacija, Volume 57 Research Papers Issue 4, November 2024 argued that there is a possibility of moderate multicollin- lower layers, known as “synaptic weights.” The output earity if VIF>5 or TOL<0.2 (Bahovec & Erjavec, 2009). of each layer is governed by the activation function em- Since each of these indicators has certain advantages and ployed, and the most effective active function suggested is disadvantages, they should both be examined to conclude the sigmoidal function (Karlik & Olgac 2011). Therefore, whether multicollinearity exists. this work employs the ANN to train and evaluate the the- Multicollinearity often appears in empirical research, oretical model, quantifying the importance of independent especially in regression models. Although there is no exact variables. solution for multicollinearity, independent variables that contribute to it may be excluded from the model (Kvan- li et al., 2003). It is important to emphasize that VIF and 5 Results TOL only indicate that the model is not ideal (Kvanli et al., 2003; O’Brien, 2007). 5.1 Step 1: Multiple regression analysis A two-step approach for assessing the proposed study model has been established in previous research (Ster- Based on all the information and inputs above, the nad Zabukovšek et al., 2019). To evaluate the relevance MLA has the following form: of the constructs in the proposed conceptual model, an GE = 0.283 + 0.60 * TC + 0.033 * EPI + 0.115 * PS importance-performance map analysis was employed. + 0.645 * RQ + 0.134 * GTA + ε (1) Furthermore, we investigated and verified the impact of For the full model, all five variables were included in independent factors on dependent variables using artificial the model to suppose that they influence the government’s neural networks (ANN), a computerized method used to effectiveness. The MLA results are presented in Table 3. estimate complex and non-linear features of interactions W for the model. In contrast, the type of contract (TC) between variables. Research by Alhumaid et al. (2021) variable was not shown as statistically significant since its proposes that an ANN has three separate modalities: trans- p-value was greater than 0.05 (p-value = 0.376). The co- fer function, network design, and learning rule. To be more efficient of determination (R-square) is high (R2=0.869), precise, these modalities may be classified as feed-forward indicating that the model fits the data well. This means that multilayer perceptron (MLP) networks, radian bases, and 86.9% of the variance of the dependent variable govern- convolutional networks. A widely used approach is the ment effectiveness (GE) is explained by the inclusion of Multilayer Perceptron (MLP) network, comprising layers four independent variables (PS, RQ, GT, EPI, GTA, and of inputs and outputs linked by hidden nodes. The input TC). layer of a neural network transfers unprocessed data to the Table 3: Variables in the full model Variables Coefficients Standard errors t-values p-values Hypothesis Conclusion Constant 0.283 0.142 1.995 0.048* PS 0.115 0.054 2.107 0.037* H1  (+5%) RQ 0.645 0.063 10.239 <0.001** H2  (+1%) EPI 0.033 0.011 2.896 0.004** H3  (+1%) GTA 0.134 0.050 2.679 0.008** H4  (+1%) TC 0.060 0.068 0.887 0.376 H5 ∅ Note: ** statistically significant at 1%; * 5; Source: Authors’ work 371 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 4: Correlation matrix GE PS RQ EPI GTA TC GE Pearson’s r 1.000 PS Pearson’s r 0.718** 1.000 RQ Pearson’s r 0.919** 0.698** 1.000 EPI Pearson’s r 0.280** -0.011 0.226 1.000 GTA Pearson’s r 0.210* 0.281** 0.277 -0.109 1.000 TC Pearson’s r 0.426** 0.410** 0.462 0.248 0.443 1.000 Source: Authors’ work Source: Authors’ work Figure 3: Heatmap of the correlations between dependent and independent variables Table 5: Tolerance (TOL) and the variance inflation factor (VIF) Variables TOL VIF PS 0.466 2.146 RQ 0.437 2.290 EPI 0.799 1.251 GTA 0.740 1.351 TC 0.617 1.621 Source: Authors’ work 372 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Results indicate that an increase in regulatory quality Multicollinearity was also tested with tolerance and within a particular country will directly increase govern- the variance inflation factor (VIF). As shown in Table 5, ment effectiveness, which is in line with part of Hypothesis tolerance was higher than 0.1, and VIF was lower than 10 1, defining that regulatory quality has a positive impact on for all variables included in the model. Based on the re- government effectiveness. Besides, an increase in govern- sults, it can be concluded that the model has no multicol- ment take attractiveness within a particular country will di- linearity problem. rectly increase government effectiveness, which is in line Additionally, the normality of residuals was test- with the part of Hypothesis 2, defining that government ed using the Kolmogorov-Smirnov, Shapiro-Wilk, and take has a positive impact on government effectiveness. An Jarque-Bera tests. The results suggest that the empirical increase in the intensity of petroleum exploration and pro- significance level for the Kolmogorov-Smirnov test is duction activities within a particular country will directly 0.200, the Shapiro-Wilk test is 0.786, and the Jarque-Bera increase government effectiveness, which is in line with test is 0.829. Accordingly, it can be concluded that at the part of Hypothesis 3, defining that the intensity of petro- significance level of 0.05, the null hypothesis stating that leum exploration and production activities has a positive residuals are normally distributed cannot be rejected. impact on government effectiveness. Finally, an increase The model diagnostics have shown that each prob- in political stability within a particular country will direct- lem’s analysis fulfilled the basic model assumptions and ly increase government effectiveness, which is in line with proved that the initial assumptions were not undermined. part of Hypothesis 3, defining that political stability has The model diagnostics have shown that each prob- a positive impact on government effectiveness. However, lem’s analysis fulfilled the basic model assumptions and due to the fact that the type of contract (TC) variable did proved that the initial assumptions were not undermined. not enter into the reduced model due to its insignificance to the full model, the part of Hypothesis 5 defining that the 5.2 Step 2: Artificial neural network type of contract used within the particular country does not influence government effectiveness was confirmed. analysis The correlation matrix and associated parameters are presented in Table 4 to test for the possible presence of The ANN was conducted using JASP 0.19. The same multicollinearity. Among independent variables, the max- variables were used as in the MLA. However, only those imum linear correlation of 0.698 was observed between variables that were significant in the MLA were used for PS and RQ, while the coefficients of the linear correlations training in the development of the ANN, indicating that TC among other variables were lower. Since all Pearson’s co- was discarded. efficients among independent variables are less than 0.9, In the current study, the logistic sigmoid function aids we can conclude that there is no multicollinearity problem in activating both output and hidden neurons with algo- in the model. Figure 3 presents the heatmap of the correla- rithm Rprop- (Resilient Propagation), which is a gradi- tions between dependent and independent variables. ent descent-based optimization algorithm primarily used Source: Authors’ work Figure 4: Neural network training graphs 373 Organizacija, Volume 57 Research Papers Issue 4, November 2024 for training ANN (Figure 4a). It is a variant of the Rprop The model summary for the Neural Network Regres- algorithm that modifies the weight updates by adapting sion in Table 6 provides key metrics for evaluating the the step sizes based on the sign of the partial derivatives model’s performance. The network consists of 4 hidden of the loss function. Unlike traditional gradient descent, layers with 18 nodes each, and the data is split into training Rprop- ignores the magnitude of the gradient, focusing (n=105), validation (n=12), and test (n=13) sets. The mod- instead on its sign to decide the direction of the update, el is optimized based on the validation set’s mean squared making it effective for handling vanishing gradients and error (MSE), which is 0.200, while the test set MSE is improving convergence speed in training deep ANN (Igel slightly higher at 0.344. et al., 2005). In order to reduce overfitting in the ANN, In the present study, the sigmoid function stimulates we employed cross-validation techniques with a ratio of the activity of both output and hidden layers. The an- 90:10:10 for testing, training, and validating the collect- alytical method was employed to determine the optimal ed data. The ANN models exhibit relative errors of 0.200 number of concealed layers, which was calculated to be 1 and 0.344 for training data and testing data, respectively (Figure 5). Furthermore, we employed a cross-validation (Figure 4b). These results suggest that the optimum ideal approach to assess and train the collected data to prevent number of layers for the models is 4. Based on the minimal overfitting in the ANN (Ahmed et al., 2021). rise in relative errors to testing from the training dataset, in conjunction with the use of ANN, it can be inferred that the suggested research models exhibit higher efficiency. Table 6: Model Summary: Neural Network Regression Hidden Layers Nodes n(Train) n(Validation) n(Test) Validation MSE Test MSE 4 18 105 12 13 0.200 0.344 Note: The model is optimized with respect to the validation set mean squared error. Source: Authors’ work Source: Authors’ work Figure 5: Network Structure Plot 374 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 7: Feature importance metrics Mean dropout loss Variable 25 permutations 50 permutations 100 permutations RQ 1.724 1.735 1.732 PS 0.631 0.635 0.641 EPI 0.609 0.612 0.613 GTA 0.559 0.558 0.555 Source: Authors’ work Source: Authors’ work Figure 6: Predictive Performance Plot Table 7 presents the feature importance metrics, rep- 6 Conclusion resented by the mean dropout loss, indicating the relative significance of each variable in the model. A lower mean The changing relationships in global petroleum mar- dropout loss suggests that the variable is more important to kets during the late 20th century and the increase and fluc- the model’s predictive capability. In this table, regulatory tuation of petroleum prices in the early 21st century have quality (RQ) is of the highest importance, suggesting it is increased the economic importance of revenue and profit the most crucial variable. Political Stability (PS) follows from their production. As a result, the legal relationships while Government Take Attractiveness (GTA) has the low- in petroleum exploration and production processes and the est importance. Different levels of permutations revealed regulation of these relationships through state interven- similar results, indicating the stability of the solution. tions have gradually achieved a universal value and be- In addition, we examined the prediction performance come a crucial subject within the competence of legislative plot to validate the neural network’s computational effi- and executive government authority. ciency and precision. The neural network model of the The objective of the study is to analyze the determi- data produced a Root Mean Square value of 0.344 and a nants that impact the effectiveness of government oper- coefficient of determination of 0.891. These values outper- ations in the petroleum sector. The present study utilizes formed those obtained from MLS, suggesting that ANN MLA to investigate the possible influence of political sta- has considerable potential for analyzing government ef- bility, regulatory quality, the level of petroleum explora- fectiveness in the petroleum sector. Figure 6 verifies a tion and production activities, government take, and con- well-defined correspondence between the observed and tract type on government performance in the petroleum anticipated values of both models. 375 Organizacija, Volume 57 Research Papers Issue 4, November 2024 sector. Moreover, ANN was investigated to ascertain the ness in the petroleum sector. The neural network model, importance of independent variables. featuring four hidden layers and 18 nodes, demonstrated The empirical research indicated that political stabil- that regulatory quality and political stability were the most ity influences government effectiveness in the petroleum significant variables, with the highest predictive capabili- sector and additionally stressed the importance of politi- ty. The neural network’s performance, as indicated by the cal stability in developing the national petroleum industry. validation and test mean squared errors, suggests a robust Regulatory quality was shown to be another factor influ- model that complements the findings of the MLA. encing government effectiveness in the petroleum sector, While the analysis provided significant insights, it is thus proving that petroleum legislation is the main factor important to acknowledge certain limitations. The relative- in regulating the complex relationships between govern- ly small sample size may restrict the generalizability of the ments and oil companies in petroleum exploration and pro- findings, and the complexity of the neural network model duction activities. poses a risk of overfitting despite the use of cross-valida- Accordingly, the empirical research showed that the tion techniques. Additionally, the study primarily focused intensity of petroleum exploration and production activi- on a specific set of variables, potentially overlooking other ties influences government effectiveness in the petroleum factors that might influence government effectiveness in sector and thus demonstrated that countries with a higher the petroleum sector. intensity of petroleum exploration and production activi- Future research should consider expanding the da- ties also have greater government effectiveness in the pe- taset to include a broader range of countries and varying troleum sector. The correlation between government take economic contexts, which could enhance the model’s ro- in terms of attractiveness ranking and government effec- bustness and applicability. Further exploration of alterna- tiveness verified that countries with a more attractive gov- tive machine learning techniques, such as deep learning ernment take and the fiscal regime has better government or ensemble methods, could provide deeper insights and effectiveness. However, empirical research has shown that improve predictions’ accuracy. Additionally, incorporat- the type of contract does not influence government effec- ing external factors, such as global oil market dynamics tiveness in the petroleum sector. and technological advancements in petroleum extraction, Hypothesis H1-H4 was accepted since the model in- could offer a more comprehensive understanding of gov- dicated that government effectiveness is positively in- ernment effectiveness in the petroleum sector. fluenced by the country’s political stability, regulatory quality, intensity of petroleum exploration and production activities, and government take attractiveness. In contrast, Acknowledgement hypothesis H5 was rejected due to the lack of a relationship between government effectiveness and type of contract. This paper is the result of the project “Entrepre- One of the most important examinations was the em- neurship and management in modern business” UNIN- pirical confirmation that the type of contract used when DRUŠ-24-1-3 of the University North, Croatia. awarding petroleum exploration and production rights to oil companies does not influence government effective- ness. The theory presented suggested that generalizations Literature are often made about the superiority of a concessionary system over a production-sharing system from the oil com- Ahmed, S. R., Kumar, A. K., Prasad, M. S., & Keerthi- pany’s point of view, despite the overwhelming similari- vasan, K. (2021). Smart IOT based short term ties from the economic, accounting, and financial points forecasting of power generation systems and quality of view, suggesting that the choice of system may not be improvement using resilient back propagation neural such a critical issue. The theory suggested that neither type network. Revista geintec-gestao inovacao e tecnolo- of contract, concession (royalty and tax based) contract gias, 11(3), 1200-1211. or production sharing contract, is better nor worse, as, Alhumaid, K., Habes, M., & Salloum, S. A. (2021). from the economic perspective, the same objectives can Examining the factors influencing the mobile learning be achieved. The empirical research demonstrated that the usage during COVID-19 Pandemic: An Integrated SEM-ANN Method. IEEE Access, 9, 102567-102578. type of contract did not influence government effective- https://doi.org/10.1109/access.2021.3097753 ness in the petroleum sector, thus further supporting the Azcel, A., & Sounderpandian, J. (2009). Complete Busi- presented theory. This is one of the most important the- ness Statistics. New York: The McGraw-Hill. oretical contributions since the theory to date has specu- Bahovec, V., & Erjavec N. (2009). Introduction to Econo- lated based on various observations without quantification metric Analysis (in Croatian). Zagreb: Element. through empirical research. Belsey, D. A., Kuh, E., & Welsch, R. E. (2004). Regres- The neural network analysis provided valuable in- sion diagnostics: Identifying influential data and sights into the key determinants of government effective- sources of collinearity. New York: Wiley. 376 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Brunnschweiler, C. N., & Bulte, E. H. (2008). The re- Governance. Colorado School of Mines, Working source curse revisited and revised: A tale of paradoxes Paper No. 2016-02. Retrieved from http://econbus. and red herrings. Journal of Environmental Econom- mines.edu/working-papers/wp201602.pdf ics and Management, 55(3), 248-264. https://doi. Kvanli, A. H., Pavur R. J. & Keeling K. B. (2003). org/10.1016/j.jeem.2007.08.004 Introduction to Business Statistics. United States: Doric, B. (2017). A sustainable administrative design for Shepherd, Inc. petroleum sector performance management. Doctoral Markus, U. (2014). Oil and Gas: The Business and Poli- dissertation. University of Ljubljana. tics of Energy. United States: Palgrave Macmillan. Falola, T., & Genova, A. (2005). The Poli- O’Brien, R. M. (2007). A Caution Regarding Rules of tics of the Global Oil Industry. https://doi. Thumb for Variance Inflation Factors. Quality and org/10.5040/9798400698781 Quantity, 41 (5), 673–690. Green, J., & Smith, A. (2023). Energy Law in a Nutshell. Parra, F. (2004). Oil Politics: A Modern History of Petro- West Academic Publishing. leum. London: IB Tauris. Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. Seba, R. D. (2008). Economics of worldwide petroleum C. (1995). Multivariate Data Analysis (3rd ed). New production (3rd ed.). Tulsa: PennWell. York: Macmillan. Shaffer, B. (2011). Energy Politics. Philadelphia: Univer- Hammerson, M. (2011). Upstream Oil and Gas: Cases, sity of Pennsylvania Press. Materials and Commentary (1st ed.). London: Globe Simkins, B., & Simkins, R. (2013). Energy Finance: Law and Business. Analysis and Valuation, Risk Management, and the Heller, P. R. P., & Marcel V. (2012). Institutional De- Future of Energy. United States: Wiley. sign in Low-Capacity Oil Hotspots. Retrieved from Šošić, I. (2004). Applied statistics (in Croatian). Zagreb: http://www.resourcegovernance.org/sites/default/ Školska Knjiga. files/documents/institutional-design-in-low-capaci- Sternad Zabukovšek, S., Kalinic, Z., Bobek, S., & Tom- ty-oil-hotspots.pdf inc, P. (2019). SEM-ANN based research of factors’ Holden, S. (2013). Avoiding the resource curse the case impact on extended use of ERP systems. Central Norway. Energy Policy, 63, 870-876. https://doi. European Journal of Operations Research, 27(3), org/10.1016/j.enpol.2013.09.01 703-735. https://doi.org/10.1007/s10100-018-0592-1 Igel, C., Toussaint, M., & Weishui, W. (2005). Rprop Tabachnick, B. G., & Fidell, L. S. (2001). Using Mul- using the natural gradient. Trends and Applications tivariate Statistics (4th ed.). Boston, MA: Allyn and in Constructive Approximation, 259-272. https://doi. Bacon. org/10.1007/3-7643-7356-3_19 Thurber, M. C., Hults, D. R., & Heller, P. R. P. (2011). IHS Energy. (2016). IHS Petroleum Economics and Exporting the “Norwegian Model”: The effect of ad- Policy Solutions - Complete Country E&P Informa- ministrative design on oil sector performance. Energy tion and Petroleum Sector Risk Analysis. Available at: Policy, 39(9), 5366-5378. https://doi.org/10.1016/j. https://cdn.ihs.com/www/pdf/PEPS-brochure.pdf enpol.2011.05.027 IHS Energy (2016a). Petroleum Economics and Policy Thurman, H. V. (2022). The Legal Landscape of Oil and Solutions (PEPS). Retrieved from https://www.ihs. Gas. Cambridge University Press. com/products/oil-gas-petroleum-news-peps.html World Bank. (2016). Worldwide Governance Indicators. Johnston, D. (1994). International Petroleum Fiscal Sys- Retrieved from http://info.worldbank.org/governance/ tems Analysis (1st ed.). Tulsa: PennWell Corporation. wgi/index.aspx#home Johnston, D. (2000). Fiscal System Design: The Ideal System. Petroleum Accounting and Financial Man- agement Journal, 19(2), 121-127. Johnston, D. (2003). International Exploration Econom- Barbara Dorić a scientist (PhD in economics) who ics, Risk, and Contracts Analysis. Tulsa: PennWell worked for Croatian most established companies in Corporation. the last 10 years. She received her PhD education at Karlik, B., & Olgac, A. V. (2011). Performance analysis the University of Ljubljana, Faculty of Economics. She of various activation functions in generalized MLP has published several scientific papers in the field of architectures of neural networks. International Jour- economics and has participated in several international nal of Artificial Intelligence and Expert Systems, 1(4), conferences. In her career, she was Executive Director 111-122. of the Centre for Monitoring of Energy Business and Kaufman, R. K., Dees, S. & Sanchez, M. P. (2004). Does Investments, President of the Management Board of OPEC Matter? An Econometric Analysis of Oil Pric- Croatian Hydrocarbon Agency, Managing Director at es. The Energy Journal, 25 (4), 68. LNG Croatia, Member of the Management Board at Kemal, M. (2016). Ownership Rights versus Access INA Group, etc. Rights Allocation to Critical Resources: An Empiri- cal Study of the Economic Impact of Changes in Oil 377 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Dinko Primorac is a Croatian scientist, university Mirjana Pejić Bach is a full professor at the professor and entrepreneur. He received his Department of Informatics, Faculty of Economics undergraduate education at Webster University, in Zagreb. She holds a PhD in system dynamics graduated from the Faculty of Economics in Zagreb, modelling from the Faculty of Economics, University and received his PhD from Megatrend University. of Zagreb. She was trained at the MIT Sloan School As a professor, he has taught at several Croatian of Management in system dynamics and OliviaGroup and international universities. He has published in data mining. Mirjana is the leader and collaborator numerous scientific and professional articles, as of numerous projects in which she cooperates with well as university textbooks. He has participated in Croatian companies and international organizations, numerous international scientific conferences, and as mainly through European Union projects and the a reviewer, he gives his contribution to professional bilateral research framework. Her research areas are and scientific journals. As a scientist, he participates in the strategic application of information technology in several domestic and international economic scientific business, data science, simulation modelling, research projects. His specialities are entrepreneurship methodology, qualitative and quantitative, especially and macroeconomics. He is a member of several multivariate statistics and modelling structural supervisory boards of renowned Croatian companies. equations. Učinkovitost vlade v naftnem sektorju: dvostopenjska analiza, ki združuje linearno regresijo in umetne nevronske mreže Ozadje in namen: Za spodbujanje razvoja naftne industrije mora država vzpostaviti regulativni okvir, ki standardizi- ra naložbene pogoje. Cilj raziskave je bil raziskati dejavnike učinkovitosti vlade v naftnem sektorju. Zasnova/metodologija/pristop: Izvedena je bila večkratna regresijska analiza, da bi raziskali, ali na učinkovitost vlade v naftnem sektorju vplivajo politična stabilnost države, regulativna kakovost, intenzivnost dejavnosti razi- skovanja in proizvodnje nafte, vladni prevzem in vrsta uporabljene pogodbe. Dodatno smo izvedli analizo umetne nevronske mreže, da bi ugotovili pomembnost neodvisnih spremenljivk. Rezultati: Politična stabilnost, regulativna kakovost, privlačnost vlade in intenzivnost naftnih dejavnosti pozitivno vplivajo na učinkovitost vlade. Privlačnejši državni prevzem poveča učinkovitost, medtem ko vrsta pogodbe o do- delitvi pravic do nafte ni bistveno vplivala na učinkovitost. Analiza umetne nevronske mreže je pokazala, da sta bili najpomembnejši spremenljivki regulativna kakovost in politična stabilnost. Zaključek: Raziskava je pokazala, da so politična stabilnost, zakonodajna kakovost in intenzivnost naftnih dejavno- sti ključni dejavniki za izboljšanje učinkovitosti vlade v naftnem sektorju. Te ugotovitve imajo praktične posledice, saj poudarjajo pomen stabilnega in dobro reguliranega okolja za doseganje večje učinkovitosti vlade v naftni industriji. To oblikovalce politik in strokovnjake v panogi opremi z uporabnimi vpogledi za izboljšanje uspešnosti sektorja. Ključne besede: Energetska politika, Učinkovitost vlade, Uspešnost naftnega sektorja, Upravljanje z naftnimi viri, Razvoj industrije 378 Organizacija, Volume 57 Research Papers Issue 4, November 2024 DOI: 10.2478/orga-2024-0027 Using the Nominal Group Technique to Design an Appropriate Communication Strategy Among Strategic Procurement and Internal Stakeholders Božidar LENARČIČ1, Armand FAGANEL2 1 Faculty of Organisation Studies, Slovenia, bozidar.lenarcic@gmail.com 2 University of Primorska, Faculty of Management, Slovenia, armand.faganel@gmail.com Background/Purpose: The implementation process of the procurement strategy requires consensus among in- ternal stakeholders. An effective procurement strategy aligns with specific business requirements and accommo- dates the individual introversion of internal stakeholders. From the latter, we formulate our research question: How is the classification of internal procurement stakeholders designed in the validation of the procurement strategy? The research objective is to present the result in the form of a schematic description and a recommendation for a communication strategy, based on insights from existing theoretical knowledge and the formulation of professional references. Methods: Nominal Group Technique (NGT) with 13 participants has been applied to analyze viewpoints on pro- curement functions. To ensure reliability and validity, Cronbach’s alpha test, Spearman coefficient, and t-tests have been used. Results and Conclusion: Study categorizes internal procurement stakeholders into four groups based on their stance towards strategy implementation: positive, cautiously positive, negatively independent, and negative stake- holders. For each group, we recommend a tailored communication strategy that includes elements such as inform- ing, guiding, motivating, and providing credible data. The research results offer important insights into the dynamics among organizational stakeholders and how communication impacts the success of strategy implementation. We have found that under-standing and adapting to various types of stakeholders are crucial for success, from those highly involved and positively inclined to those with the potential to oppose or hinder implementation. Employing specific communication strategies for each group can significantly enhance the chances of successful implementa- tion and sustainable execution of strategies. Keywords: Procurement strategy, Internal stakeholder management, Communication strategy, Nominal Group Tech- nique, Internal communication 1 Introduction expectation of receiving reciprocal value in return. Stra- tegic procurement, a vital but relatively under-researched Supplier marketing is the practice of organizations ac- field, entails systematic approaches aligning with an or- tively interacting with suppliers, fostering robust supplier ganization’s goals, encompassing sourcing, contract man- relationships, and generating value for suppliers with the agement, and supplier relations (O’Brien, 2019). It focuses 1 Received: 17th November 2023; Accepted: 24th May 2024 379 Organizacija, Volume 57 Research Papers Issue 4, November 2024 on long-term planning, risk management, and value crea- tion strategies is vital. tion, enhancing efficiency and competitive edge (Tuncel In strategy implementation, category managers must et al., 2020). Our study addresses this research gap by use innovative communication methods, underpinned by proposing a novel, practical model for classifying inter- research and analyses of procurement category charac- nal stakeholders in strategic procurement during strategy teristics. As category management encompasses a broad validation and devising effective communication strategies organizational scope, it’s vital to include stakeholders in (O’Brien 2019). Central to strategic procurement is the de- the communication plan. This ensures that crucial messag- velopment and implementation of procurement strategies, es related to the category project are effectively conveyed with strategy validation as a crucial step to align internal (Voegeli & Finger 2021). stakeholder or customer consensus, recognizing their di- Effective management of procurement strategies verse assumptions impacting strategy execution (O’Brien, hinges on accurately identifying and understanding inter- 2019). Our research focuses on identifying methods to nal stakeholders and their interests. In strategy formula- classify internal stakeholders in specific organizations, en- tion, it’s essential to categorize stakeholders by their power suring alignment with the organization’s overall procure- and interest, assessing their potential support or opposition ment strategy, and determining effective communication (Cordell & Thompson, 2018). Internal stakeholders, rang- (Xie et al., 2020). approaches to gain their consensus. We ing from individuals to groups within the organization, aim to clarify stakeholder classification (Murray-Webster often have diverse interests and perspectives. Recogniz- & Simon, 2006) from a procurement perspective, present- ing these differences (Bisung & Dickin, 2019) and creat- ing a schema for strategic procurement professionals. This ing synergy (Dimitrova & Gallucci, 2017) among various schema is based on stakeholder interest in strategies, fa- stakeholders is key to achieving consensus and success in cilitating the management of communication during the strategy implementation. strategy validation process (Ginige at al., 2018). Effective communication planning with internal stake- 2.2 The classification of internal holders is vital (Rowlinson, et al. 2009) in validating pro- curement strategies. It involves a structured and planned stakeholders approach to disseminating both specific and general infor- mation across various areas. The process begins with iden- Effective management of internal stakeholders, through tifying internal stakeholders, followed by selecting appro- understanding their interests and implementing clear ac- priate communication methods for engaging with them tion plans, is essential for successful procurement strategy (Ortbal at al., 2016). We recommend a stepwise approach, implementation. This involves a critical analysis phase to illustrated in the process diagram provided, for classifying identify and assess each stakeholder’s influence and im- internal stakeholders from most opposing to most support- portance in strategy execution. Category managers play a ive. Based on classification Ginige at al., (2018) tailored key role in recognizing and addressing diverse stakeholder communication strategies (García, 2012) are then recom- requirements, with the goal of minimizing conflicts, obsta- mended. cles, and risks (Jilani, 2018) thereby facilitating effective collaboration and achieving organizational objectives. 2 Literature Review Involving internal stakeholders is key for building trust, crucial for effective communication and mutual un- derstanding, thereby enhancing strategy success through 2.1 Procurement Strategy their knowledge and experience (Savage et al., 1991). Yet, Implementation caution is necessary, as stakeholders wield power and can impact strategy implementation positively or negatively Strategic planning (Dimitrova & Gallucci 2017) in or- (Ginige at al., 2018). ganizations is crucial particularly for managing procure- Classifying internal stakeholders is vital for creating a ment strategies (O’Brien, 2019). where the validation of useful stakeholder list, assessing their impacts, and manag- the strategy is a key phase. This process involves a range ing them effectively to achieve organizational objectives. of procurement scenarios and diverse stakeholders with This process includes identifying, understanding, and cate- varying interests, which can sometimes conflict. These gorizing stakeholders for successful collaboration in activ- conflicts, when managed effectively, can enhance efficien- ity implementation. Rowlinson et al. (2009) describes this cy and spur innovation (Chinyio & Olomolaiye 2009). as involving information gathering, examining stakeholder A significant challenge in this process is communication interests, strategy definition, behavior prediction, and im- planning. It is often mistakenly assumed that all stakehold- plementing a communication strategy. ers will interpret messages uniformly (Ke, 2011). There- Literature review reveals that effective stakeholder fore, ensuring precision and assertiveness in communica- understanding can be enhanced by using data in various forms, including graphical and visual representations10. 380 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Various methods have been proposed for classifying in- The overarching consensus underscores the vital role of ternal stakeholders, including categorization based on risk active stakeholder management in the successful execu- and cooperation potential (Mitchell et al., 1997). Then tion of procurement strategies, fostering understanding, classification by influence power and importance of activi- building trust, minimizing conflicts, and nurturing long- ties, development of comprehensive processes from stake- term beneficial relationships. holder identification to communication strategies (Turner Recognizing internal stakeholders is key for under- & Jain, 2002). Fletcher et al., (2003) proposed emphasis standing their interests and in-fluence on strategy imple- on stakeholder management in relation to hierarchical mentation (Ginige, Amaratunga and Haigh, 2018). The structure and operational scope. Appropriate classification first step, as advised Chinyio & Olomolaiye (2009). is methods are essential for effective internal stakeholder to create a comprehensive and clear list of stakeholders, management and control. which is foundational for subsequent analysis. Prioriti- The 1997 model by Mitchell, Agle, and Wood classi- zation and ranking them in a table or graphic provide an fies stakeholders based on power, legitimacy, and urgency. overview of key stakeholders. Identifying intersections Power refers to the ability to influence the organization, and potential conflicts among stakeholders, as suggest legitimacy is the perceived validity of their claim, and ur- Ortbal at al., (2016) is critical in stakeholder management. gency is the need for immediate attention (Wood, Mitchell, Developing effective communication strategies is cru- Agle & Bryan, 2021). cial for addressing the challenges inherent in managing The Stakeholder Circle methodology, is introduced by internal stakeholders, particularly in complex scenarios Bourne and Walker (2006) as a process for categorizing marked by diverse interests. The significance of managing internal stakeholders that includes identification, prioriti- internal stakeholders is further highlighted by its role in zation, visualization, strategy development, and commu- understanding their concerns and expectations, dissem- nication effectiveness monitoring. While categorization inating information, and identifying uncertainties. These often uses specific terms, caution is advised against labels elements collectively enhance internal relationship man- like “time bomb” or “dangerous” due to potential subjec- agement and overall process efficiency (Voegeli & Finger, tivity and risks (Chinyio & Olomolaiye 2009). Stakeholder 2021). analysis, an important management element (Raum 2018), Stakeholder analysis extends beyond influence and in- helps identify stakeholders with significant impacts and in- terest to include power, support, and relationship, visually forms engagement strategies (Ginige at al., 2018). represented for clarity on maps. Stakeholders are classified Effective stakeholder management in procurement ne- into categories based on these dimensions. Key Stakehold- cessitates multiple aspects: identifying key stakeholders, ers, with high power and interest, greatly influence imple- understanding their interests, leveraging their knowledge, mentations and include executives and advisors. Satisfied preparing for challenges, considering their needs, and Stakeholders, often regulatory bodies, have high pow- implementing engagement strategies. Managing internal er but low interest and need periodic updates. Informed stakeholders effectively involves recognizing their distinct Stakeholders, like local or community groups, show high requirements and using a 2 x 2 matrix based on power and interest but low power, requiring regular communication interest for classification. This includes strategies like pri- and consideration. Low-Impact Stakeholders, with mini- oritizing collaboration for high power, high interest stake- mal power and interest, have a limited impact on risks, but holders, ensuring satisfaction for high power, low interest their management is crucial if risks increase (Chinyio & ones, monitoring needs of low power, low interest stake- Olomolaiye 2009). holders, and regularly updating those with low power but Difference between internal (primary) and external high interest. Continuously updating this matrix is crucial (secondary) stakeholders based on their connection to a for identifying stakeholders with potential unexpected project. Internal stakeholders are directly involved, often demands, ensuring successful execution of procurement contributing financing and having a contractual connec- strategies (Chinyio & Olomolaiye 2009). tion, while external stakeholders, though affected, are not Authors like Raum (2018), emphasize the importance critical to the project’s survival. Stakeholders are catego- of analyzing internal stakeholders in strategic manage- rized as supportive, neutral, or unsupportive, with efforts ment, focusing on their roles, influence, and potential made to shift unsupportive stakeholders towards neutrality conflicts. Effective stakeholder management (Chinyio & or support (Rowlinson, et al. (2009). Olomolaiye 2009) involves key aspects such as inform- For successful procurement strategy implementation, ing stakeholders, achieving commercial objectives, con- stakeholder involvement is key, aligning their interests trolling their influence, and incorporating their opinions in with strategy outcomes. This involves preparing a stake- decision-making. This encompasses the identification of holder list and assessing their potential impact on the pro- stakeholders, understanding their influence and interests, ject. Stakeholders are classified by power and interest, with establishing effective communication, managing expecta- tools like classification maps aiding in identifying various tions, and involving them in decision-making processes. types, from blockers or critics to advocates or supporters 381 Organizacija, Volume 57 Research Papers Issue 4, November 2024 (Ginige at al., 2018). 2.3 Communication strategies with The influence of internal stakeholders is linked to their internal stakeholders power, stemming from their capacity to persuade or co- erce. Hester uses fuzzy logic to categorize stakeholder atti- tudes on a scale, assigning weights to these attitudes to fa- The critical aspects of “strategic communication,” cilitate the transformation of non-supportive stakeholders where “strategic” indicates deliberate, planned activities, into supporters (Chinyio & Olomolaiye 2009). and “communication” significantly affects an organiza- Identifying key stakeholders, whether individuals or tion’s strategic goals (García, 2022). Strategic commu- groups, is crucial in stakeholder categorization (Chinyio nication management involves systematic planning and & Olomolaiye 2009). Once stakeholders are identified, the implementation of information flow through selected me- next step is to define activities and integrate them based on dia, targeting specific messages to appropriate audiences their levels of support. Different models for categorizing at the right times (Chinyio & Olomolaiye 2009). Effec- internal stakeholders facilitate their identification, clas- tive stakeholder involvement and tailored communication sification, and determination of appropriate engagement strategies are essential for achieving organizational goals. activities. Continuous monitoring of their power and in- Building trust through effective communication among terests is essential for effectively managing potential neg- stakeholders is vital, as is maintaining a shared focus on ative impacts (Chinyio & Olomolaiye 2009). Caution is organizational goals. Communication takes various forms, advised in labelling stakeholders with terms like “Sleeping including verbal, non-verbal, written, and digital channels. Giant” or “Time Bomb,” as these can be counterproductive Organizational communication flows through downward, and sensitive, particularly if stakeholders have access to upward, and horizontal pathways. Downward communica- the categorization model (Ginige at al., 2018). Ensuring tion flows from top to middle management, then to subor- controlled access, careful handling, and maintaining con- dinates. Upward communication channels feedback from fidentiality are crucial due to the sensitive nature of the lower-level employees to top management, while horizon- categorization model (O’Brien, 2019). tal communication occurs among leadership, employees, Key principles for managing internal stakeholders owners, and department heads. Effective communication include actively monitoring their interests and demands, requires proper timing, simplicity, clarity, appropriateness, respecting organizational procedures and stakeholder op- credibility, and style. Regular meetings and presentations, erations, identifying interdependencies, ensuring fair dis- underpinned by thorough analyses, aid in effective com- tribution of benefits and burdens, collaborating to mitigate munication. Building relationships among stakeholders risks, being aware of potential conflicts of interest, and through regular interviews is crucial for successful en- avoiding activities that could jeopardize human rights or gagement (Chinyio & Olomolaiye 2009). increase risks. Effective conflict management is crucial Gong emphasizes the necessity of fact-based commu- for successful strategy implementation. Emphasizing open nication strategies, balancing between direct and indirect communication and thorough reporting is vital, alongside approaches, and managing risks systematically through pre-assessing stakeholders’ needs and expectations. Build- identification, analysis, and prioritization. Understanding ing positive interpersonal relationships fosters trust, col- the current situation is essential before developing strate- laboration, and reduces conflict risks, preventing delays gies that align with stakeholders’ needs and interests. In or- and additional costs. Early identification of potential ac- ganizations, stakeholder relationships involve key figures tions, tensions, or issues is key to managing them effec- who can influence organizational goals. Internal stakehold- tively, ensuring timely responses and safeguarding strat- ers, characterized by their influence, legitimacy, and own- egy or project execution (Chinyio & Olomolaiye 2009). ership, include diverse groups affecting the organization’s The literature underscores the criticality of strategic strategy. Stakeholder management is pivotal, requiring the planning in procurement, focusing on category managers’ identification and understanding of stakeholders’ unique roles, their innovative communication methods, and stake- demands and rights. Gathering and analyzing information holder involvement. It emphasizes the need to identify and is crucial for effective collaboration and support in strategy analyze internal stakeholders, considering their influence execution (Chinyio & Olomolaiye 2009). and importance for successful strategy execution. Stake- Internal stakeholders are instrumental in executing pro- holder involvement is key to building trust and enhanc- curement strategies, advocating for interests and collective ing strategy effectiveness. Analyzing stakeholders’ power commitment to projects (Månsson, 2018). Differentiating and interest through classification matrices is crucial, as between internal stakeholders involved in organizational are communication strategies in stakeholder management. activities and external stakeholders affected by these ac- Continuous monitoring and conflict management through tivities is essential (Jilani, 2018). Effective interpersonal open communication are vital for maintaining harmonious communication is key to aligning and harmonizing stake- stakeholder relationships (O’Brien, 2019). holders’ interests and perceptions (Chinyio & Olomolaiye 2009). 382 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Gaining internal stakeholder support and consensus is strategies aligned with organizational objectives. It under- essential for successful procurement strategy implemen- scores the importance of various organizational commu- tation. This requires strategies that positively influence nication pathways, such as downward, upward, and hori- stakeholders and minimize adverse effects. Developing a zontal, stressing the need for clear, timely, and credible comprehensive communication plan is crucial for consen- interactions. The critical role of internal stakeholders in sus-building. Stakeholder communication management in- executing procurement strategies is emphasized, requiring volves identifying assumption discrepancies, understand- comprehensive identification, understanding, and manage- ing strategy requirements, analyzing stakeholders’ needs ment of their diverse interests, expectations, and influences and interests, and evaluating the effectiveness of commu- on projects. Effective communication is key, with tactical nication strategies. Effective communication with internal relationship management recommended for its efficacy. stakeholders is vital to harmonize relationships and align Furthermore, the literature suggests customized commu- objectives. Understanding the significant impact of stake- nication strategies like Collaboration, Defense, Involve- holders on strategy execution is key. Employing tactical ment, and Monitoring to manage stakeholders according to relationship management tools, such as matrices and cate- different risk levels and interests, underlining the necessity gorizations, aids in effective communication. Acknowledg- of flexible communication methods for varied stakeholder ing and responding appropriately to the diverse influences contexts. and demands of internal stakeholders is crucial for effec- tive strategy management and implementation. Authors Chinyio V Olomolaiye (2009) outlined four tailored com- 3 Materials and Methods munication strategies for stakeholder management based on risk levels and stakeholder interest. The Collaboration 3.1 The survey samples Strategy focuses on stakeholder involvement in high-risk, high-interest scenarios to reduce risks and increase sup- Our research targets procurement specialists, specifi- port. The Defence Strategy, for high-risk, low-interest sit- cally in the strategic procurement sector. Given the limited uations, manages conflicts to maintain stability, building size of the professional community in this field, we fo- trust through clear strategic communication. The Involve- cused on selecting experts with a minimum of three years ment Strategy, suitable for low-risk, low-interest contexts, of procurement experience, at least five years of overall maintains relationships through continuous communica- work experience, and an age threshold of 25 years or old- tion and information exchange. The Monitoring Strategy, er. These selected experts, drawn from medium-sized and applied in low-risk, high-interest situations, emphasizes large companies, were chosen based on criteria specified in closely monitoring stakeholders’ interests and needs to the Companies Act of the Republic of Slovenia (ZGD-1), keep them informed and engaged. Article 55. The literature emphasizes strategic communication and stakeholder management within organizations, high- lighting the need for deliberate, fact-based communication Figure 1: The approach to implementing the Nominal Group Technique 383 Organizacija, Volume 57 Research Papers Issue 4, November 2024 3.2 Nominal Group Technique FRIEND: Low power but high interest and positive, serving as key advisors. Our research utilized the NGT to explore stakeholder REBEL: Influential with high interest but negative at- classification in strategic procurement, involving 19 stra- titudes, needing strategic management. tegic procurement employees responsible for strategy de- OBSTRUCTOR: Low power, highly obstructive, ne- sign and stakeholder collaboration. The NGT’s systematic cessitating involvement despite challenges. approach enabled us to gain insights into current classifi- SUPPORTER: Influential and interested, albeit some- cation practices within modern organizations. This quali- times passive, requiring active engagement. tative method, initially developed for healthcare (in 1960s) SPECIALIST: Low power and interest but significant emphasizes individual perspectives and achieves consen- in attitude, needing basic engagement. sus through a structured process, blending interview and BOMB: Powerful but low-interest, with a negative ap- focus group elements. The NGT process, effective both proach requiring careful mitigation. in-person and online, involves participants expressing OPPORTUNIST: Low energy and interest, demanding their views anonymously, thus encouraging open discus- cautious monitoring. sion without bias (Vogelnik, 2016; Jones & Hunter, 1995; SAFEGUARD: Interested and positive but indecisive, Mullen et al., 2021; Mason at al., 2021). needing extra analytical verification. The online approach (Fig. 1) allowed us to efficient- TACTICIAN: Moderate interest in decision-making, ly conduct our research while ensuring broad participant requiring engagement stimulation. engagement. The facilitator introduces the topic, followed Participants rated these stakeholders on a scale from 1 by individual idea contribution and group discussions (Vo- to 5 based on their support level. After classifying and dis- gelnik, 2016). Participants then rank these ideas, with the cussing these stakeholders, the final step involved reaching facilitator consolidating and presenting the final rankings. group consensus on the descriptions and communication In our study, we adapted NGT for an online format using strategies, with the results incorporated into the Results MS Teams, ensuring inclusivity and wider participation, as chapter. outlined in the research procedure. Confirmation for participation was received from 14 3.3 Reliability and validity of the method candidates via email, followed by coordinating a meeting on MS Teams, with consideration for COVID-19 pan- In our research within the strategic procurement func- demic guidelines. On the meeting day, 13 participants at- tion, we conducted a measurement to ensure the validity, tended, aligning with the recommended focus group size adhering to the principle of measuring precisely what we (Vogelnik, 2016), 9 to 12 experts. Though smaller groups intended (Vogelnik, 2016). We identified and defined in- are recommended for more in-depth discussions, with suc- ternal stakeholders, akin to types like Sleeping Giant and cessful applications using 4 to 7 participants (Chiwire et Time Bomb. Similar to Hester’s approach, we categorized al., 2022). We began with a 10–15-minute casual conver- stakeholder relationships on a 5-point scale from extreme- sation to explain NGT procedures and foster a comfortable ly interested too uninterested. To confirm the reliability of communication atmosphere, presenting the NGT execu- our measurement questions, we employed the Cronbach tion processes and research purpose, particularly focusing alpha test (Cronbach, 1951), assessing the reliability coef- on our research question. During the meeting, we intro- ficient as advised by relevant authors (Stadler et al., 2021; duced descriptions similar like Ginige at al., (2018) of in- Diedenhofen & Much, 2016). ternal stakeholders, such as “Sleeping Giant” and “Time The Cronbach alpha reliability method used also Črv, Bomb”. Assessing the relationship (Hester, 2015) between (2023); Ball, (2029); Martín Martínez, and Vela Llauradó, stakeholders’ interest and power on a 5-point scale, we (2022) is crucial in verifying the consistency of question- evaluated them from extremely interested to disinterest- naire measurements. The alpha coefficient (α) ranges from ed, aligning with the described stakeholder types. We then 0 to 1, with values closer to 1 indicating higher internal applied four communication strategies for stakeholder consistency. A Cronbach alpha value above 0.7 suggests management as per established procedures (Chinyio & Ol- good construct reliability (Diedenhofen & Much, 2016) omolaiye 2009). denoting the reliability of the questionnaire’s questions. In our study, we employed the NGT to categorize in- We used descriptive statistics to present subject data, in- ternal stakeholders in strategic procurement. The process, cluding respondents’ age, total work experience, and pro- lasting about 50 minutes, involved outlining procurement curement experience, demonstrating the participants’ ex- strategy management, guiding participants to identify and pertise. Spearman’s correlation coefficient was employed discuss internal stakeholders, and consolidating ten unique to determine the strength of the correlation between var- stakeholder profiles. These profiles include: iables, indicating a stronger association (Statstutor, 2019; MENTOR: Strong, influential, and highly interested, Yan et al., 2029). requiring careful attention. 384 Organizacija, Volume 57 Research Papers Issue 4, November 2024 To check the statistical properties of the data, we used a procurement function. Our study’s participant group con- t-test and assessed statistical significance at a significance sisted of eight males and five females, ranging in age from level of 0.05. A value of (p) less than 0.05 is considered 26 to 58, with an average age of 46.8 years. They had an statistically significant (Yan et al., 2029). Statistical signif- average service length of 24.4 years, including an average icance was calculated using the equation: of 15.7 years in procurement. The reliability of our questionnaire, as indicated by a Cronbach’s alpha score of 0.709, falls within the “good” range (Plešnik, 2019; Šćepanović 2020), suggesting a high level of confidence in its reliability. This score, on a scale where 1 represents perfect reliability, confirms the depend- We adopted four communication strategies for stake- ability of our research instrument. holder management, akin to researchers Chinyio & Olo- To further validate our findings, we employed the molaiye (2009). After finalizing the list, we categorized Spearman correlation coefficient (rs), which ranges be- stakeholders by priority and interest, using tables or graphs tween -1 and +1 (Statstutor, 2019). A near-perfect correla- as per author recommendations, addressing our research tion of rs = 0.997 was observed between age and total work question. experience. Additionally, strong correlations were found for both total work experience and procurement function 4 Results experience (rs = 0.664), and between age and procurement function experience (rs = 0.666). We also conducted a statistical t-test to observe p-val- Using the NGT, we gathered and organized data ues against a significance threshold of 0.0539. Our analy- through a structured process, guided by a facilitator. We sis revealed statistical significance (p) between respond- systematically arranged the data into a table, presenting ents’ age and total work experience (8.75-14), as well as the descriptions of internal stakeholders, their commu- between total work experience and procurement function nication recommendations, and their levels of interest in experience (1.33-02) and between respondents’ age and strategy implementation. These levels ranged from most procurement function experience (1.30-02). These find- to least supportive, shedding light on the key aspects of ings further reinforce the reliability of the data collected classifying internal stakeholders in a strategically oriented from the respondents. Table 1: Demographic data of respondents Respondent sex age Total work experience Total years in procurement 1 m 44 18 18 2 m 48 25 15 3 m 57 36 32 4 w 49 27 18 5 m 45 21 6 6 w 46 22 12 7 m 58 37 37 8 m 32 7 3 9 m 59 39 6 10 m 55 34 26 11 w 58 38 21 12 w 26 5 4 13 w 32 8 6 385 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 2: Display of Nominal Group Technique Results Table 3: Descriptions of internal stakeholders and communication recommendations NAMING STAKE- ESCRIPTION COMMUNICATION RECOMMENDATIONS LEVEL OF HOLDERs D INTEREST High-status leader with significant GODFATHER influence and a supportive stance Proactively engage and tailor communication toward the project. to their needs to maintain support. 60 Often a project leader, shows high SUPPORTER interest and positivity, though may be Activate their support through engagement passive. and motivation. 56 FRIEND Positively inclined with low organizational power, like a colleague. Serve as advisors and information sources 48 SPECIALIST Key for their technical expertise despite Keep informed and consult as technical limited power. advisors. 45 Focused on analytical, data-driven SAFEGUARD decisions, with a generally positive Back strategies with data and analysis, and outlook. value their verification advice. 40 TACTICIAN Plays a critical role in decisions but has Align project goals with their interests to an unclear position. boost engagement. 38 OPPORTUNIST Shows low interest, engaging in minor Communicate with caution and understanding tasks. to avoid negative impacts. 29 BOMB In leadership but opposes the project. Identify early and plan strategic interventions to limit their hindrance. 27 REBEL High leadership, opposes project. Utilize influential supporters to diminish their opposition. 21 OBSTACLE Middle management that actively seeks Illustrate the project’s organizational benefits to impede the project. and manage their involvement carefully. 15 386 Organizacija, Volume 57 Research Papers Issue 4, November 2024 We analyzed our results using the Nominal Group strates the consensus-based classification of internal stake- Technique (NGT), as shown in Table 1. From 13 partic- holders in strategic procurement during the validation of a ipants, we collected 10 names of internal stakeholders, procurement strategy. It details stakeholders’ relationships, ensuring comprehensive understanding. Table 1 presents from most to least supportive, and specifies communica- demographic data of respondents involved in a research tion strategies for low-importance procurement catego- study focusing on procurement experts. The table outlines ries. This reflects a modern organizational emphasis on the sex, age, total work experience, and years in procure- fostering individual autonomy, accountability, and active ment for each of the 13 participants. Among these partici- engagement through critical analysis. pants, there are eight males (m) and five females (w), with ages ranging from 26 to 59 years. The total work expe- rience among the respondents varies from 5 to 39 years, 5 Discussion indicating a wide spectrum of professional backgrounds. The years spent specifically in procurement also show di- The discussion around the classification and man- versity, ranging from 3 to 37 years. This variation in de- agement of internal stakeholders in procurement strategy mographic and professional backgrounds of the respond- validation has highlighted significant insights. It emerged ents provides a comprehensive view of the experience and from the dialogues with participants that internal stake- perspectives within the field of procurement, facilitating a holders are dynamic entities; their initial positions may robust analysis of internal stakeholder dynamics based on shift, indicating the fluid nature of stakeholder attitudes the substantial and varied experiences of the participants. and the importance of continuous engagement and mon- Descriptions were added to these names for clarity itoring. This adaptability is particularly crucial for groups among participants. Each participant rated the stakehold- with an ambivalent stance towards the issue, underscoring ers’ interest levels individually using a Likert scale from the need for tailored communication strategies to address 1 to 5. their specific needs and concerns. In our study, we utilized descriptive statistics to ana- The Nominal Group Technique (NGT) was chosen lyze the diverse opinions of procurement experts on in- for its structured process, rich data generation, and con- ternal stakeholders. The findings revealed a spectrum of sensus-building capabilities. NGT ensures equal partic- interest levels, with “obstacle” garnering the least and ipation, minimizing bias, and involves independent idea “godfather” the most interest. Notably, significant shifts in generation, group discussion, and ranking, ideal for com- interest were observed, particularly from “opportunist” to plex issues like stakeholder classification. NGT produces “tactician”, indicating a need for increased motivation dur- prioritized ideas, aiding decision-making and ensuring ing such transitions. Conversely, minor shifts from “bomb” strategies are accepted and supported. However, it can be to “opportunist” and from “tactician” to “safeguard” sug- time-consuming and relies on skilled facilitation. It may gested low participation interest but high obstruction ten- not delve as deeply into underlying reasons as other quali- dency, and a neutral yet positive stance towards strategy, tative methods. Despite these limitations, NGT’s strengths respectively. These insights are crucial for understanding in structured, inclusive discussions make it effective for stakeholder dynamics in strategic procurement. classifying and managing internal stakeholders in strategic Statistical measures like variance (σ²), standard devia- procurement. tion (σ), and median (M) re-veal the diversity in subjects’ In the context of previous studies, our findings reso- views on internal stakeholder strategy interest. High- nate with study Månsson (2018), who emphasized the crit- er variance and standard deviation signify more diverse icality of the credibility of information in synthesizing per- opinions. The highest dispersion is seen in “bomb” and spectives from various internal stakeholders. This aspect “safeguard” (σ² = 0.533), and the lowest in “obstacle” (σ² underlines the importance of not only understanding the = 0.130). The highest deviation (σ = 0.76) is also observed varied positions of stakeholders but also ensuring the relia- in “bomb” and “safeguard”, while the lowest is in “obsta- bility of the information exchanged, which forms the basis cle” (σ = 0.38). Median values (M) indicate the majority for effective communication and strategy implementation. interest rating for each stakeholder. This analysis clarifies Our research aligns with previous studies (Chinyio & the varied perceptions and opinions among respondents re- Olomolaiye 2009) in stakeholder management, particular- garding different internal stakeholders. ly in developing strategies based on stakeholders’ power Table 2 outlines internal stakeholders’ descriptions, and interest. Yet, our study stands out by tailoring com- characteristics, and communication method recommenda- munication strategies to stakeholders’ attitudes, organizing tions, providing a clear understanding of interaction com- them into distinct categories such as positive, undecided plexities in strategic procurement. This aids in managing positive, negative independent, and outright negative. internal stakeholders and validating procurement strate- Like earlier research Chinyio & Olomolaiye (2009), we gies. emphasize the vital role of ongoing engagement and moni- Table 3 resulting from the NGT’s final step, demon- toring of stakeholders in executing successful procurement 387 Organizacija, Volume 57 Research Papers Issue 4, November 2024 strategies. Distinguishing our work from Chinyio & Olo- tation within organizations. It stands out by categorizing molaiye (2009), is our approach and depth in analyzing stakeholders into four groups based on their attitudes to- stakeholders. While prior research predominantly uses a wards procurement strategy, thereby offering managers a power-interest grid, our study delves into stakeholders’ at- refined, adaptable framework for customizing communi- titudes toward strategy execution, offering a richer, more cation and leadership strategies, leading to more efficient intricate understanding of their attitudes and behaviors. and sustainable strategy implementation. This innovative This method recognizes the complexities in stakeholder approach enhances the effectiveness of stakeholder en- dynamics that go beyond simple power and interest dy- gagement, responding to the dynamic nature of stakehold- namics. Our focus on rigorous data collection methods er attitudes and is applicable across various countries and enhances the validity of our findings, demonstrating a de- industries, addressing universal dynamics. Our study not tailed exploration of stakeholder dynamics. However, this only contributes to stakeholder management literature but approach necessitates regular reassessment of stakeholder also provides practical tools for managers, supported by attitudes, diverging from the more stationary power-in- robust statistical validation using Cronbach’s alpha and terest analysis. Overall, our research, alongside previous Spearman’s correlation coefficients, and further reinforced studies, significantly advances the field of stakeholder by t-tests. This approach, which aligns communication management in procurement, with our unique emphasis strategies with distinct stakeholder groups, adds a novel on attitudes providing deeper insights into the nuances of dimension to stakeholder management, underlining the stakeholder relationships intricate relationship between stakeholder attitudes and ef- Our study introduces a straightforward method for fective communication. stakeholder categorization, enabling managers to swiftly This study’s limitations stem from its narrow sample identify and implement suitable communication strategies. and focus on private sector procurement within a particular This approach, focused on tailoring communication to nation, potentially affecting the universality of its findings. each stakeholder group, aligns stakeholder attitudes direct- The specific context may not reflect the varied dynamics of ly with communication methods. In contrast, other models larger, diverse populations or other organizational environ- like Murray-Webster & Simon (2006), may necessitate ments, and differences in cultural, economic, and regula- more in-depth analysis, rendering them more appropriate tory aspects could impact the efficacy of the strategies and for detailed stakeholder management scenarios. In sum- stakeholder categorizations proposed. Additionally, the mary, our research offers a simpler method for stakeholder fluid nature of communication in stakeholder management categorization and communication based on attitude. Oth- suggests that strategies successful in our context may not er studies (Murray-Webster & Simon, 2006) provide more translate universally. Consequently, the study’s insights, elaborate frameworks, encompassing power, interest, and while valuable for a specific national context, have limited attitude, suitable for more detailed applications in stake- broader application. Future research should include longi- holder management. tudinal and cross-industry studies in diverse cultural set- Our study enhances the understanding of stakeholder tings to broaden stakeholder management understanding, management in procurement by offering a detailed per- develop globally applicable communication models, and spective on internal stakeholder dynamics. We categorize contribute to establishing international standards in this stakeholders into distinct groups - positive, undecided pos- domain. itive, negative independent, and negative - each with tai- lored communication strategies. This approach illuminates the complex and diverse attitudes of stakeholders, aiding Literature in crafting adaptable and effective communication meth- ods. Nevertheless, the fluid nature of stakeholder positions Ball, H.L. (2019), ‘Conducting Online Surveys’, and our focus on procurement limit the generalizability of Journal of Human Lactation, 35(3), 413–417. our findings, indicating the need for regular strategy reas- DOI: 10.1177/0890334419848734 sessment and broader applicability. Future research should Bisung, E. and Dickin, S. (2019), ‘Concept mapping: En- delve into how stakeholder attitudes evolve over time and gaging stakeholders to identify factors that contribute to the long-term effectiveness of these communication strat- empowerment in the water and sanitation sector in West egies in varied organizational contexts, thereby enriching Africa’, SSM - Population Health, 9(September 2019). DOI: 10.1016/j.ssmph.2019.100490 the overall understanding of stakeholder management in Bourne, L. and Walker, D.H. (2006), ‘Visualizing different business sectors. stakeholder influence—Two Australian exam- ples’, Project Management Journal, 37(1), 5–21. 6 Conclusion DOI: 10.1177/875697280603700102 Chinyio, E. and Olomolaiye, P. (2009), ‘Construction stakeholder management’, John Wiley & Sons, New Our research marks a significant advancement in un- York, USA. derstanding stakeholder dynamics for strategic implemen- 388 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Chiwire, P., Evers, S.M., Mahomed, H. and Hiligsmann, 104. M. (2022), ‘Identification and Prioritization of At- Månsson, P. (2018), ‘Mapping challenges and op- tributes for a Discrete Choice Experiment Using the portunities for aggregating information on sys- Nominal Group Technique: Patients’ Choice of Public temic risks from multiple stakeholders’, Pro- Health Facilities in Cape Town, South Africa’, Value cedia Engineering, 212(2017), 736–743. in Health Regional Issues 27, 90–98. DOI: 10.1016/j. DOI: 10.1016/j.proeng.2018.01.095 vhri.2021.06.005 Martín Martínez, L. and Vela Llauradó, E. (2022), Companies Act – ZGD-1. Official Gazette of the Republic ‘Questionnaire on the Training Profile of a Learn- of Slovenia, No. 42/06 of 19 April 2006). Available at: ing Therapy Specialist: Creation and Validation http://www.pisrs.si/Pis.web/pregledPredpisa?id=ZA- of the Instrument’, Sustainability, 12(21), 1–20. KO4291#, last accessed on 03.12.2023. DOI: 103390/su12219159 Cordell, A. and Thompson, I. (2018), ‘The Category Man- Mason, S., Ling, J., Mosoiu, D., Arantzamendi, M., Tserkezo- agement Handbook’, Routledge, New York, USA. glou, A.J., Predoiu, O. and Payne, S. (2021), ‘Undertak- DOI: 10.4324/9781351239585 ing Research Using Online Nominal Group Technique: Cronbach, L.J. (1951), ‘Coefficient alpha and the internal Lessons from an International Study (RESPACC)’, structure of tests’, Psychometrika 1951, 16, 297–334 Journal of Palliative Medicine, 24(12), 1867–1871. Črv, S. (2023), ‘Načela menedžmenta sistemov kakovosti DOI: 10.1089/jpm.2021.0216 kot gradniki uspešnosti gospodarskih družb’, Doctor- Mitchell, R.K., Agle, B.R. and Wood, D.J. (1997), ‘Toward al dissertation. Faculty for organization studies, Novo a theory of stakeholder identification and salience: De- mesto, Slovenia. fining the principle of who and what really counts’, Diedenhofen, B. and Musch, J. (2016), ‘cocron: A Web In- Academy of management review 22(4), 853–886. terface and R Package for the Statistical Comparison Mullen, R., Kydd, A., Fleming, A. and McMillan, L. (2021), of Cronbach´s Alpha Coefficients’, International Jour- ‘A practical guide to the systematic application of nom- nal of Internet Science, 11(1), 51–60. inal group technique’, Nurse researcher, 29(1):14–20. Dimitrova, V. and Gallucci, T. (2017), ‘Reverse Logistics DOI: 10.7748/nr. 2021.e1777 and Circular Economy-Working Together? Izvestia’, Murray-Webster and R., Simon, P. (2006), ‘Making sense Journal of the Union of Scientists-Varna, 2(Economic of stakeholder mapping’, PM World today, 8(11), 1–5. Sciences Series), 13–20. O’Brien, J. (2019), ‘Category Management in Purchasing: Fletcher, A., Guthrie, J., Steane, P., Roos, G. and A Strategic Approach to Maximize Business Profitabil- Pike, S. (2003), ‘Mapping stakeholder per- ity’, 4th ed. Kogan Page: London, UK. ceptions for a third sector organization’, Jour- Ortbal, K., Frazzette, N. and Mehta, K. (2016), ‘Stakeholder nal of intellectual capital, 4(4), 505–527. journey mapping: An educational tool for social entre- DOI: 10.1108/14691930310504536 preneurs’, Procedia Engineering, 159(June), 249–258. García, C. (2012), ‘Using strategic communi- DOI: 10.1016/j.proeng.2016.08.170 cation for nation-building in contemporary Plešnik, J. (2019), ‘Pripadnost kot gradnik modela voden- Spain: The Basque case’, International Jour- ja zaposlenih v osnovnih šolah’, Doctoral dissertation. nal of Strategic Communication, 6(3), 212–231. Faculty for organisation studies, Novo mesto, Slove- DOI: 10.1080/1553118X.2012.678523 nia. Ginige, K., Amaratunga, D. and Haigh, R. (2018), Raum, S. (2018), ‘A framework for integrating systematic ‘Mapping stakeholders associated with so- stakeholder analysis in ecosystem services research: cietal challenges: A Methodological Frame- Stakeholder mapping for forest ecosystem services work’, Procedia Engineering, 212, 1195–1202. in the UK’, Ecosystem Services, 29, 170–184. DOI: DOI: 10.1016/j.proeng.2018.01.154 10.1016/j.ecoser.2018.01.001 Hester, P. (2015), ‘Analyzing Stakeholders Using Fuzzy Cog- Rowlinson, S., & Tuuli, M. (2009). Stakeholder man- nitive Mapping’, Procedia Computer Science, 61, 92–97. agement through relationship management. In Per- DOI: 10.1016/j.procs.2015.09.159 formance Improvement in Construction Management Jilani, P.A. (2018), ‘Indirect Procurement Strategies for (187-207). Routledge. Supply Chain Sustainability’, Doctoral dissertation, Savage, G.T., Nix, T.W., Whitehead, C.J. and Blair, Walden University, Walden. J.D. (1991), ‘Strategies for assessing and man- Jones, J. and Hunter, D. (1995), ‘Consensus aging organizational stakeholders’, Acade- methods for medical and health services re- my of management perspectives, 5(2), 61–75. search’, Qualitative Research, (311), 376–380. DOI: 10.5465/ame.1991.4274682 DOI: 10.1136/bmj.311.7001.376 Stadler, M., Sailer, M. and Fischer, F. (2021), ‘Knowledge Ke, G. (2011), ‘Cultural difference effects on business: as a formative construct: A good ala is not always bet- Holding up Sino-U.S. Business Negotiation as a mod- ter’, New Ideas in Psychology, 60(August 2020), 1–4. el/les effets de la difference culturelle sur les affaires: DOI: 10.1016/j.newideapsych.2020.100832 En prenant la négociation sino-américaine comme un Statstutor. Spearman’s correlation, (2019). Available modèle. Cross-Cultural Communication, 7(2), 101– at: https://www.statstutor.ac.uk/resources/uploaded/ 389 Organizacija, Volume 57 Research Papers Issue 4, November 2024 spearmans.pdf, last accessed on 03.12.2023 and optimizing procurement processes. With over two Šćepanović, D. (2020), ‘Gradniki modela nacionalnega decades of diverse experience, including roles in Spain. programa primarne preventive urinske inkontinence v Sloveniji’, Doctoral dissertation. Faculty for organisa- tion studies, Novo mesto, Slovenia. Armand Faganel is teaching marketing courses and Tuncel, E., Kong, D.T., McLean Parks, J. and van acting as vice-dean at the University of Primorska, Kleef, G. A. (2020), ‘Face threat sensitivity in dis- Faculty of Management (UP FM) since 2003. Before tributive negotiations: Effects on negotiator self-es- that, he gained work experience in sales and marketing teem and demands’, Organizational Behavior and at different international companies for 13 years. His Human Decision Processes, 161(July), 255–273. research mainly covers diverse marketing and selling DOI: 10.1016/j.obhdp.2020.07.004 topics, service quality perception, sustainable tourism, Turner, R. and Jain, A. (2002), ‘Agile meets CMMI: Cul- etc. He has been granted and has directed several ture clash or common cause?’, In Conference on Ex- international and state-funded research projects on treme Programming and Agile Methods. Springer Ber- marketing and selling politics and practice. He has lin, Heidelberg, Germany, 153–165. been awarded two times the University’s recognition Voegeli, G. and Finger, D.C. (2021), ‘Conflicting interests for pedagogical excellence. over hydropower: Identifying and representing stake- holder perspectives on new projects using causal map- ping’, EGU General Assembly Conference Abstracts, 19–30. Voegeli, G. and Finger, D.C. (2021), ‘Disputed dams: Mapping the divergent stakeholder per- spectives, expectations, and concerns over hy- dropower development in Iceland and Switzer- land’, Energy Research and Social Science, 72(4). DOI: 10.1016/j.erss.2020.101872 Vogelnik, M. (2016), ‘Simulacijska tehnika skupinske di- namike, nova kvalitativno-kvantitativna tehnika’, ba- chelor thesis. Univerza v Ljubljana, Ljubljana, 2016. Wood, D. J., Mitchell, R. K., Agle, B. R., & Bryan, L. M. (2021). Stakeholder identification and sa- lience after 20 years: Progress, problems, and prospects. Business & Society, 60(1), 196–245. DOI: 10.1177/0007650318816522 Xie, Y., Li, Z., Qi, J., Zhang, K., Zhang, B. and Qi, F. (2020), ‘Optimal video communication strategy for intelligent video analysis in unmanned aerial vehicle applications’, Chinese Journal of Aeronautics, 33(11), 2921–2929. DOI: 10.1016/j.cja.2020.06.009 Yan, Z., Wang, S., Ma, D., Liu, B., Lin, H. and Li, S. (2019), ‘Meteorological factors affecting pan evapo- ration in the Haihe River Basin, China’ Water, 11(2), 1–18. DOI: 10.3390/w11020317 Yu, Z., Guindani, M., Grieco, S. F., Chen, L., Holmes, T. C. and Xu, X. (2022), ‘Beyond t test and ANOVA: applications of mixed-effects mod- els for more rigorous statistical analysis in neu- roscience research’, Neuron 2022, 110(1), 21–35. DOI: 10.1016/j.neuron.2021.10.03 Božidar Lenarčič is a Purchasing Category Manager at a company in Novo Mesto, Slovenia, and serves as a lecturer at the Faculty of Organization Studies in Novo Mesto. He specializes in strategic purchasing, focusing on the development and implementation of procurement strategies, negotiating with key suppliers, 390 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Uporaba tehnike nominalne skupine za oblikovanje ustrezne komunikacijske strategije med strateškimi na- bavnimi in notranjimi deležniki Namen: Proces implementacije nabavne strategije zahteva doseganje konsenza med notranjimi deležniki. Učinkovi- ta strategija se usklajuje s specifičnimi poslovnimi zahtevami in hkrati upošteva edinstvene značilnosti notranjih de- ležnikov. Namen te študije je obravnavati raziskovalno vprašanje: Kako je zasnovana klasifikacija notranjih nabavnih deležnikov med validacijo nabavne strategije? Cilj je predstaviti ugotovitve v obliki shematičnega opisa in priporočil za komunikacijsko strategijo, pri čemer se opiramo na obstoječa teoretična spoznanja in strokovne reference. Metode: Študija uporablja tehniko nominalne skupine (NGT) z 13 udeleženci za analizo perspektiv glede nabavnih funkcij. Za zagotavljanje zanesljivosti in veljavnosti ugotovitev raziskava uporablja Cronbachov alfa test, Spearma- nov korelacijski koeficient in t-test. Rezultati in zaključek: Študija klasificira notranje nabavne deležnike v štiri skupine glede na njihov odnos do imple- mentacije strategije: pozitivni, previdno pozitivni, negativno neodvisni in negativni deležniki. Za vsako skupino je pri- poročena prilagojena komunikacijska strategija, ki vključuje elemente, kot so obveščanje, usmerjanje, motiviranje in zagotavljanje verodostojnih podatkov. Raziskava ponuja pomembne vpoglede v dinamiko med organizacijskimi de- ležniki in vpliv komunikacije na uspešnost implementacije strategije. Razumevanje in prilagajanje različnim vrstam deležnikov je ključno za uspeh, od tistih, ki so močno vključeni in pozitivno naravnani, do tistih, ki lahko nasprotujejo ali ovirajo implementacijo. Uporaba specifičnih komunikacijskih strategij za vsako skupino lahko bistveno poveča verjetnost uspešne in trajnostne izvedbe strategij. Ključne besede: Strategija nabave, Upravljanje notranjih deležnikov, Komunikacijska strategija, Tehnika nominalne skupine, Notranja komunikacija 391 Organizacija, Volume 57 Research Papers Issue 4, November 2024 DOI: 10.2478/orga-2024-0028 The Effect of Brand Association and Brand Image on the Brand Awareness of Lithuanian Companies of Digital Products Gita ŠAKYTĖ-STATNICKĖ Klaipėdos valstybinė kolegija, Higher Education Institution, Klaipėda, Lithuania, g.statnicke@kvk.lt Purpose: Brand awareness enables brand recognition in different market conditions, which is very important for companies selling digital products or providing digital services. Purpose of this paper is to determine the effect of brand association and brand image on the brand awareness of Lithuanian companies of digital products. These factors are important and have an impact on brand awareness, but their impact on the brand awareness of digital products in particular has not been investigated. Methods: A questionnaire was conducted to a sample of 403 consumers who buy and use digital products in Lith- uania. The Multiple Linear Regression (MLR) model was created and a statistically significant direct positive impact of brand association and brand image on brand awareness of companies of digital products was determined. Results: The empirical research of the effect of brand association and brand image on the brand awareness of Lith- uanian companies of digital products revealed a statistically significant direct positive impact of brand association and brand image on brand awareness of companies of digital products. The strongest direct positive impact of brand image on brand awareness than brand association. Conclusion: The findings of research provide valuable insights into the crucial role played by brand association and brand image in building brand awareness for companies of digital products. Keywords: Brand Awareness, Brand Association, Brand Image, Digital Products 1 Introduction maintaining strong brand awareness can lead to increased customer trust, loyalty, and a competitive edge, enabling Brand awareness is one of the most important factors digital product companies to attract and retain customers effectively. Effective digital marketing strategies, consist- for companies of digital products, as it influences con- ent branding, and engaging content play pivotal roles in sumer recognition and familiarity with their brands in the enhancing brand awareness for companies offering digital highly competitive digital marketplace. Brand awareness products, fostering lasting connections with their target au- is closely related to brand recognition and consumer mem- dience and driving business growth. Brand awareness in- ories of the brand (Ramadani and Rachmawati, 2022), dicates that consumers can identify and remember a brand, which is very important in the digital space with high com- which establishes a foundation for future long-term com- petition between digital products or services. Building and mitment between consumers and brands (Liu et al., 2020). 1 Received: 20th May 2024; Accepted: 29th August 2024 392 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Organizations are facing frustration in measuring brand 2.2 Link between Brand Association and perceptions like brand awareness (Phulpoto et al., 2018) Brand Awareness of Companies of because the market of digital products has its own specif- ics and with the rapid growth of this market, not enough Digital Products research has been done on brand awareness of companies of digital products. Although new digital products can Brand association is important to the customer, espe- penetrate the market much more quickly which can raise cially when there are too many products on the market, brand awareness much faster than for physical products as digital products or digital services proliferate. Brand (Basu and Muylle, 2023), ensuring a high brand awareness association helps the customer identify the digital product is quite difficult due to the large number of competitors. brand they need that best meets their needs and expecta- If the research carried out on the market of usual physical tions (Shamsudin et al., 2020). The link between brand products shows that brands association for many products association and brand awareness in companies of digital is similar for consumers and non-customers of a product products lies in the power of consumer perceptions and especially for mobile phones (Phulpoto et al., 2018), then associations (Das, 2020; Supiyandi et al., 2022). in order to identify and improve the brands association of Positive brand associations established through effec- companies of digital product, many additional challenges. tive marketing campaigns and positive customer experi- Researchers studying brand awareness of usual physical ences contribute significantly to the overall brand aware- products indicate that there are two main constructs that ness of companies offering digital products, as consumers are most important in improving brand awareness: these are more likely to identify and engage with brand they hold are brand association and brand image, which have strong favorable associations with (Wirtz et al., 2013; Schivinski links with brand awareness (Shabbir et al., 2017; Bilgin, et al., 2016; Hamzah et al., 2021). Results show that there 2018; Bernarto et al., 2020; Liu et al., 2020; Abbas et al., is correlation between brand awareness, brand association 2021). and brand loyalty (Phulpoto et al., 2018). Considering the aforementioned relevance of brand Companies of digital products can strategically lever- awareness for companies of digital products and the prob- age brand association to enhance brand awareness by cre- lems arising from it, the aim of this paper is to determine ating meaningful connections with their target audience, the effect of brand association and brand image on the thereby solidifying their position in the competitive digital brand awareness of Lithuanian companies of digital prod- sector. ucts. In the light of the mentioned explanations the follow- ing hypothesis can be proposed: H1. Brand association has positive effect on brand awareness of companies of 2 Literature Review digital products. 2.1 Brand Awareness of Companies of 2.3 Link between Brand Image and Digital Products Brand Awareness of Companies of Digital Products Brand awareness of companies of digital products re- fers to the extent to which consumers recognize and are fa- Brand image and brand awareness are two major brand miliar with a particular brand and its offerings in the digital variables that measure the effectiveness of marketing strat- marketplace. Brand awareness is the ability of a potential egies in brand management practice (Esch et al., 2006; Liu customer to recognize or remember that a brand is part of et al., 2020). a particular digital product (Ilyas et al., 2020; Mahaputra The relationship between brand image and brand and Saputra, 2021; Pina and Dias, 2021). Brand awareness awareness in digital product companies highlights the im- refers to the awareness of a brand’s existence and its as- portance of overall brand perception and visual representa- sociation with a specific product (Clarissa and Bernarto, tion in influencing consumer recognition and awareness 2022). Brand awareness enables brand recognition in dif- (Marques et al., 2020; Makrides et al., 2020; Pina and ferent market conditions, which is very important for com- Dias, 2021). panies selling digital products or providing digital servic- A well-designed and consistent brand image across all es. The higher the brand association in the digital product, digital platforms increases brand awareness by allowing the more it will be remembered by the consumer and be consumers to quickly recognize and differentiate a compa- loyal towards the brand (Sasmita and Mohd Suki, 2015). ny’s products from competitors in the digital realm. Com- Brand awareness indicates that consumers can identify panies producing digital products must invest in develop- and remember a brand, which establishes a foundation ing a positive and attractive brand image that matches the for future long-term commitment between consumers and preferences of their target audience, thereby enhancing brands (Liu et al., 2020). 393 Organizacija, Volume 57 Research Papers Issue 4, November 2024 brand awareness and creating a lasting impression in the search included 403 respondents subjected to analysis: minds of consumers (Shahid et al., 2017; Febriyantoro, 56.33% of the sample (N=227) were female; the other 2020; Zaqi and Pradini, 2022). 43.67% (N=176) were male. Respondents were divided by There are studies that examine the relationship between age as follows: 27.79% of the sample (N=112) were 18- brand image and brand awareness and it is empirically 28 years old; 34.00% (N=137) – 29-38 years old; 26.55% supported that the higher the brand awareness, the high- (N=107) – 39-49 years old; 10.67% (N=43) – 49-59 years er the brand image (Bilgin, 2018; Bernarto et al., 2020). old; 0.99% (N=4) were over 59 years old. The distribu- Brand awareness has some connection with brand image tion of respondents by position in the company was as fol- (Abbas et al., 2021). A strong connection between brand lows: business owners / directors – 10.17% (N=41), heads awareness and brand image has been established (Liu et of department – 19.11% (N=77), specialists – 44.67% al., 2020). Using structural equation modeling approach, (N=180), workers – 17.37% (N=70); “other” was indicat- brand loyalty and brand image are found to have positive ed by 8.68% (N=35) of respondents. 37.72% of the sample effects on brand awareness (Shabbir et al., 2017). (N=152) work in a Micro enterprise, 33.0% (N=133) work In the light of the mentioned explanations the follow- in a small enterprise, 20.1% (N=81) work in a Medium ing hypothesis can be proposed: H2. Brand image has enterprise, 9.18% (N=37) work in a Large enterprise. positive effect on brand awareness of companies of digital products. 3.3 Data Analysis 3 Methodology The statistical data analysis of quantitative research was performed using IBM SPSS Statistics 25.0 software. 3.1 Measures The following main data analysis methods were ap- plied in the research. Firstly, the descriptive statistics The research was carried out using a quantitative analysis of the research data was performed. Secondly, survey – a written questionnaire. The questionnaire was the Kolmogorov-Smirnov test (N=403>50) indicated that constructed on the basis of an analysis of the scientific the results have a normal distribution. The reliability of literature dealing with the three main constructs: brand the empirical research was assessed applying Cronbach’s awareness (BAw), brand association (BAs) and brand im- alpha criterion. In the next stage, the Exploratory Factor age (BI). The questionnaire consisted of three scales meas- Analysis (EFA) were performed, to check the validity of uring brand awareness, brand association, brand image the constructs. Data suitability for factor analysis is tested and a block of demographic questions. using Bartlett’s Test / Criterion of Sphericity and the Kai- A 5-item scale designed to measuring brand awareness ser-Meyer-Olkin (KMO) measure of sampling adequacy. (example of a scale item “I aware this particular product After that, the Multiple Linear Regression (MLR) model / brand that appeared in the social media”) and a 5-item was performed. Regression modeling is one of the most scale designed to measuring brand association (example of widely used statistical processes to estimate the relation- a scale item “This particular product / brand that appeared ships between dependent and independent variables (Ete- in the social media has its own personality”) were adopt- madi and Khashei, 2021). ed from Severi and Ling (2013). A 6-item scale designed The Multiple Linear Regression (MLR) model al- to measuring brand image (example of a scale item “This lows the effect of brand association and brand image on brand has a high quality”) were adopted from Ansary and the brand awareness of Lithuanian companies of digital Nik Hashim (2018). A 5-point Likert Scale is applied for products. The aim of this MLR method is to explain the measuring brand awareness, brand association and brand relationship between the dependent variable (brand aware- image, where 1 is strongly disagree and 5 – strongly agree. ness) and two explanatory variables (brand association The questionnaire contains demographic information: and brand image). The Multiple Linear Regression mod- age, gender, position in the company, company size. el (MLR) are defined by the equation (Formula 1) (Olive, 2017; Bangdiwala, 2018): 3.2 Research Sample (1) In this research, quantitative method has been used and where: research data has been obtained via online questionnaires y – dependent variable; shared on social media. A total of 403 Lithuanian residents x_i – independent variable; who buy and use digital products participated in this re- β_i – parameter; search. ε – error Demographic characteristics of the sample. The re- 394 Organizacija, Volume 57 Research Papers Issue 4, November 2024 4 Results 4.2 Normality, Reliability and Validity Test 4.1 Descriptive Statistics The Kolmogorov-Smirnov test (N=403>50) indicated Descriptive statistical analysis was conducted to ob- that the results have a normal distribution. tain an overview of the initial characteristics of the three The reliability of the empirical research was also as- constructs (brand awareness, brand association and brand sessed applying Cronbach’s alpha criterion. In the re- image) studied. Analysis includes minimum values, max- search, the scales for measuring the constructs of brand imum values, average values and standard deviations. Ta- awareness, brand association and brand image are compat- ble 1 presents the results of testing descriptive statistical ible and reliable (α=0.853, α=0.905 and α=0.941, respec- analysis. tively) (Table 2). Table 1: Descriptive Statistics (N=403) Constructs Minimum Maximum Mean Std. Deviation BAw 2.20 5.00 4.01 0.57630 BAs 2.60 5.00 3.84 0.55664 BI 2.71 5.00 4.18 0.55061 Table 2: Reliability and Validity Test (N=403) Constructs Items Rotated Component Matrixa / Cronbach’s Reliability Factor Loading Coefficients BAw1 0.733 BAw2 0.598 BAw BAw3 0.704 0.853 BAw4 0.661 BAw5 0.636 BAs1 0.718 BAs2 0.746 BAs BAs3 0.753 0.905 BAs4 0.765 BAs5 0.717 BI1 0.729 BI2 0.798 BI3 0.811 BI BI4 0.675 0.941 BI5 0.819 BI6 0.796 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 7 iterations. KMO Measure of Sampling Adequacy =0.916; p=0.000<0.05; df=630 Bartlett’s Test of Sphericity / Approx. Chi-Square =5513.157 395 Organizacija, Volume 57 Research Papers Issue 4, November 2024 To check the validity of the constructs, the Exploratory iable is between 0.617 and 0.674. The correlation coeffi- Factor Analysis (EFA) of the obtained data was performed cients of the independent variables are found to be related applying the Principal Components Analysis with Varimax to the dependent variable. rotation to determine the factor loadings. Discriminant The Variance inflation factors value VIF=1.630<4 validity was evaluated through inter-construct correlation (VIFs exceeding 4 warrant further investigation), and the coefficients. Data suitability for factor analysis is tested tolerance value is 0.614, exceeding 0.10. The results show using Bartlett’s Test / Criterion of Sphericity and the Kai- that the data variables used in this study did not have mul- ser-Meyer-Olkin (KMO) measure of sampling adequacy. ticollinearity problems. It was found that the quantitative research data are suitable for the Exploratory Factor Analysis when Bartlett’s Test 4.4 The Multiple Linear Regression of Sphericity is significant, since the KMO coefficient is 0.916 (p=0.000<0.05; df=630). By estimating the factor (MLR) Model loadings from Rotated Component Matrix the convergent validity of the measurement items was obtained and com- The regression analysis in this study is carried out us- posite reliabilities as standardized factor loadings are be- ing Multiple Linear Regression (MLR) analysis. Table 4 tween 0.598 to 0.819 which are above than suggested level presents the results on the fit of the MLR model – Model of 0.40 so they all are at significant level (Raubenheimer, Summary. R-squared (the coefficient of determination) is a 2004). statistical metric that is used to measure how much of the variation in outcome can be explained by the variation in 4.3 Correlations and Collinearity the independent variables. As can be seen from Table 4, the value of R Square is Statistics 0.518, which means that the model explains 51.8 percent of brand awareness. Before the Multiple Linear Regression analysis, we The information presented in Table 5 helps to assess performed a correlation analysis of the independent var- the statistical significance of the MLR model. The statis- iables (Table 3). tical value of F criterion is 53.651 and the model reaches The results of the correlation matrix analysis showed statistical significance with Sig.=0.000, p<0.0005. This re- that the average correlation between each independent var- gression model is indeed statistically significant. Table 3: Correlations of Constructs and Collinearity Statistics (N=403) Collinearity Statistics Constructs BAw BAs BI Tolerance VIF BAwa 1.000 Pearson BAs 0.617 1.000 0.614 1.630 Correlation BI 0.674 0.622 1.000 0.614 1.630 a. Dependent Variable: Brand Awareness p=0.000<0.05 Table 4: Model Summaryb Model R R Square Adjusted R Square Std. Error of the Estimate 1 0.719a 0.518 0.508 0.40425 a. Predictors: (Constant). Brand Image. Brand Association b. Dependent Variable: Brand Awareness 396 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Table 5: ANOVAa Model Sum of Squares df Mean Square F Sig. Regression 17.535 2 8.767 53.651 0.000b 1 Residual 16.342 100 0.163 Total 33.877 102 a. Dependent Variable: Brand Awareness b. Predictors: (Constant). Brand Image. Brand Association Table 6: Multiple Linear Regression Test Results – Coefficientsa Unstandardized Coefficients Standardized Model Coefficients t Sig. B Std. Error Beta (Constant) 0.652 0.326 1.998 0.048 1 BAs 0.334 0.092 0.322 3.637 0.000 BI 0.495 0.093 0.473 5.337 0.000 a. Dependent Variable: Brand Awareness MLR is based on the assumption that there is a line- The explanation of the Multiple Linear Regression (MLR) ar relationship between the dependent (brand awareness) equation above is as follows: and both independent variables (brand association and • The constant value of 0.652 in the equation repre- brand image). The MLR provides an indication of which sents the predicted value of brand awareness when of the variables included in the model contributed to the all the independent variables are equal to zero. prediction of the dependent variable. This requires esti- • The value of the coefficient β1 has a value of mating the Standardized Coefficients Beta. As can be seen 0.334, which means that for every one unit in- from Table 6, the highest Standardized Coefficients Beta crease in brand association, the predicted value of is 0.473 (brand image), which means that this variable brand awareness increases by 0.334 units with the explains more of the variance of the dependent variable assumption that other independent variables still (brand awareness) than brand association (Standardized have the same value. The obtained results confirm Coefficients Beta is 0.322). All variables are statistically the hypothesis H1. Brand association has positive significant, so all can be included in the equation of the effect on brand awareness of companies of digital regression model. products. According to the values of the coefficients in Table 6, • The value of the coefficient β2 has a value of 0.495 the equation of the MLR model can be constructed (For- which means that for every one unit increase in mula 2): brand image, the predicted value of brand aware- ness increases by 0.495 units with the assumption (2) that other independent variables have the same value. The obtained results confirm the hypothe- where: sis H2. Brand Image has positive effect on brand y – Brand Awareness (dependent variable); awareness of companies of digital products. x_1 – Brand Association (independent variable); x_2 – Brand Image (independent variable); ε – error. 5 Discussion Digital products often face intense competition in the market (Goyal et al., 2021; Tiwari, 2022). Strong brand 397 Organizacija, Volume 57 Research Papers Issue 4, November 2024 association helps differentiate a company’s products from tations to consider when interpreting the mentioned find- competitors, making it easier for consumers to identify and ings. The research was carried out and the Multiple Linear choose the brand they trust and prefer (Shamsudin et al., Regression analysis was performed only in Lithuania, i. 2020). When consumers have positive associations with e. the respondents only of one country participated in the a brand, they are more likely to recognize and recall the research. It is likely that cultural, economic, social, tech- brand when making purchase decisions. nological, and other differences would affect the results of The effect of the brand association variable on the research conducted in another country / region / continent. brand awareness of digital product companies shows that Thus, further research could be extended to other regions. the brand association variable has a positive coefficient re- lationship with brand awareness, suggesting that the high- er the brand association, the higher the brand awareness. Literature The results obtained in this study are in agreement with those of the results of the other authors’ research (Wirtz Abbas, U., Islam, K. M. A., Hussain, S., Baqir, M., & et al., 2013; Schivinski et al., 2016; Hamzah et al., 2021), Muhammad, N. (2021). Impact of brand image on stating that positive brand associations significantly to the customer loyalty with the mediating role of customer overall brand awareness, as consumers are more likely to satisfaction and brand awareness. International Jour- identify and engage with brands, they hold favorable asso- nal of Marketing Research Innovation, 5(1), 1-15. Doi: ciations with. 10.46281/ijmri.v5i1.987. The effect of the brand image variable on the brand Ansary, A., & Nik Hashim, N. M. H. (2018). Brand image awareness of digital product companies shows that the and equity: The mediating role of brand equity drivers brand image variable has a positive coefficient relation- and moderating effects of product type and word of ship with brand awareness, suggesting that the higher the mouth. Review of Managerial Science, 12, 969-1002. brand image, the higher the brand awareness. The results Doi: 10.1007/s11846-017-0235-2. obtained in the research agree with the results of other au- Bangdiwala, S. I. (2018). Regression: multiple lin- thors’ research, e.g. brand image is found to have positive ear. International Journal of Injury Control effects on brand awareness (Shabbir et al., 2017); brand and Safety Promotion, 25(2), 232-236. Doi: awareness has some connection with brand image (Abbas 10.1080/17457300.2018.1452336. et al., 2021); a strong connection between brand awareness Basu, A., & Muylle, S. (2023). Creating New Digital Prod- and brand image has been established (Liu et al., 2020) ucts. In Competitive Digital Innovation: Transforming and etc. Products, Processes and Business Models to Win in the The results of the study show that it is important for Digital Economy (pp. 67-87). Cham: Springer Interna- companies to actively manage their brand association and tional Publishing. brand image in order to maximize the positive effect of Bernarto, I., Berlianto, M. P., Meilani, Y. F. C. P., Mas- brand association and brand image on brand awareness. man, R. R., & Suryawan, I. N. (2020). The influence of brand awareness, brand image, and brand trust on 6 Conclusion brand loyalty. Jurnal Manajemen, 24(3), 412-426. Doi: 10.24912/jm.v24i3.676. Bilgin, Y. (2018). The effect of social media marketing The empirical research of the effect of brand associa- activities on brand awareness, brand image and brand tion and brand image on the brand awareness of Lithua- loyalty. Business & management studies: an interna- nian companies of digital products revealed a statistically tional journal, 6(1), 128-148. Doi: 10.15295/bmij. significant direct positive impact of brand association and v6i1.229. brand image on brand awareness of companies of digital Clarissa, C., & Bernarto, I. (2022). The Influence of products. Strongest direct positive impact of brand image Brand Ambassador, Brand Awareness, Brand Image on brand awareness than brand association. and Prices on Purchase Decisions on Online Market- Theoretical and practical implications. The paper is place. Business and Entrepreneurial Review, 22(2), related to a deeper understanding of the brand awareness 273-288. Doi: 10.25105/ber.v22i2.14966. of companies of digital products phenomenon and its con- Das, S. (2020). Innovations in digital banking service nections with brand association and brand image. As a brand equity and millennial consumerism. In Digital practical implication, derived from the research findings, transformation and innovative services for business formulated the MLR model equation for brand awareness and learning (pp. 62-79). IGI Global. allow digital marketing professionals to better understand Esch, F. R., Langner, T., Schmitt, B. H., & Geus, P. (2006). the influence of brand association and brand image on Are brands forever? How brand knowledge and rela- brand awareness and purposefully shape brand awareness. tionships affect current and future purchases. Journal Limitations and future research. The paper has limi- of product & brand management, 15(2), 98-105. Doi: 398 Organizacija, Volume 57 Research Papers Issue 4, November 2024 10.1108/10610420610658938. and Word of Mouth on Samsung Smartphone Purchase Etemadi, S., & Khashei, M. (2021). Etemadi multi- Decisions. JBMP (Jurnal Bisnis, Manajemen dan Per- ple linear regression. Measurement, 186, 1-21. Doi: bankan), 8(1), 73-86. 10.1016/j.measurement.2021.110080. Raubenheimer, J. (2004). An item selection procedure to Febriyantoro, M. T. (2020). Exploring YouTube Market- maximize scale reliability and validity. SA Journal of ing Communication: Brand awareness, brand image Industrial Psychology, 30(4), 59-64. and purchase intention in the millennial generation. Sasmita, J., & Mohd Suki, N. (2015). Young consumers’ Cogent Business & Management, 7(1), 1-17. Doi: insights on brand equity. International Journal of Re- 10.1080/23311975.2020.1787733. tail & Distribution Management, 43(3), 276-292. Doi: Goyal, A., Bhong, S., Kumbhare, P., & Bharadkar, R. 10.1108/ijrdm-02-2014-0024. (2021). The new era of digital marketing: a literature Schivinski, B., Christodoulides, G., & Dabrowski, D. review. PalArch’s Journal of Archaeology of Egypt/ (2016). Measuring consumers’ engagement with Egyptology, 18(10), 728-741. brand-related social-media content: Development and Hamzah, Z. L., Abdul Wahab, H., & Waqas, M. (2021). validation of a scale that identifies levels of social-me- Unveiling drivers and brand relationship implications dia engagement with brands. Journal of advertising of consumer engagement with social media brand research, 56(1), 64-80. Doi: 10.2501/JAR-2016-004. posts. Journal of Research in Interactive Marketing, Severi, E., & Ling, K. C. (2013). The mediating effects 15(2), 336-358. Doi: 10.1108/JRIM-05-2020-0113. of brand association, brand loyalty, brand image and Ilyas, G. B., Rahmi, S., Tamsah, H., Munir, A. R., & Putra, perceived quality on brand equity. Asian Social Sci- A. H. P. K. (2020). Reflective model of brand aware- ence, 9(3), 125-137. ness on repurchase intention and customer satisfac- Shabbir, M. Q., Khan, A. A., & Khan, S. R. (2017). Brand tion. Journal of Asian Finance, Economics and Busi- loyalty brand image and brand equity: the mediating ness, 7(9), 427-438. role of brand awareness. International Journal of Inno- Liu, C., Zhang, Y., & Zhang, J. (2020). The impact of vation and Applied Studies, 19(2), 416-423. self-congruity and virtual interactivity on online celeb- Shahid, Z., Hussain, T., & Zafar, F. (2017). The impact of rity brand equity and fans’ purchase intention. Journal brand awareness on the consumers’ purchase intention. of Product & Brand Management, 29(6), 783-801. Journal of Accounting & Marketing, 6(01), 34-38. Mahaputra, M. R., & Saputra, F. (2021). Relationship Doi: 10.4172/2168-9601.1000223. word of mouth, advertising and product quality to Shamsudin, M. F., Hassan, S., Ishak, M. F., & Ahmad, Z. brand awareness. Dinasti International Journal of (2020). Study of Purchase Intention towards Skin Care Digital Business Management, 2(6), 1099-1108. Doi: Products Based on Brand Awareness and Brand Asso- 10.31933/dijdbm.v2i6.1331. ciation. Journal of Critical Reviews, 7(16), 990-996. Makrides, A., Vrontis, D., & Christofi, M. (2020). The Supiyandi, A., Hastjarjo, S., & Slamet, Y. (2022). Influ- gold rush of digital marketing: assessing pros- ence of brand awareness, brand association, perceived pects of building brand awareness overseas. Busi- quality, and brand loyalty of shopee on consumers’ ness Perspectives and Research, 8(1), 4-20. Doi: purchasing decisions. CommIT (Communication and 10.1177/227853371986001. Information Technology) Journal, 16(1), 9-18. Marques, C., da Silva, R. V., Davcik, N. S., & Faria, R. Tiwari, S. P. (2022). Organizational Competitiveness and T. (2020). The role of brand equity in a new rebrand- Digital Governance Challenges. Archives of Business ing strategy of a private label brand. Journal of Busi- Research, 10(3), 165-170. Doi: 10.2139/ssrn.4068523. ness Research, 117, 497-507. Doi: 10.1016/j.jbus- Wirtz, J., den Ambtman, A., Bloemer, J., Horváth, C., res.2020.06.022. Ramaseshan, B., van de Klundert, J., Gurhan Canli, Olive, D.J. (2017). Multiple Linear Regression. In: Linear Z., & Kandampully, J. (2013). Managing brands and Regression. Springer, Cham. Doi: 10.1007/978-3-319- customer engagement in online brand communities. 55252-1_2. Journal of Service Management, 24(3), 223-244. Doi: Phulpoto, N. H., Halepoto, I. A., Hussain, A., & Ahmed, S. 10.1108/09564231311326978. (2018). Analyzing the mediating role of Brand associ- Zaqi, M., & Pradini, G. (2022). Influence of brand image ation for Brand awareness and loyalty. European Jour- and the quality of service to the decision to stay at red- nal of Business and Management, 10 (21), 121-123. dorz pondok labu. Eqien-Jurnal Ekonomi dan Bisnis, Pina, R., & Dias, Á. (2021). The influence of brand ex- 11(03), 919-929. periences on consumer-based brand equity. Journal of brand Management, 28, 99-115. Doi: 10.1057/s41262- 020-00215-5. Ramadani, A. G., & Rachmawati, E. (2022). The Influence of Brand Awareness, Brand Association, Brand Image, 399 Organizacija, Volume 57 Research Papers Issue 4, November 2024 Gita Šakytė-Statnickė, PhD in Management, Social marketing; human resource management, employees of Sciences, is an Associate Professor at Klaipėdos different generations; work engagement; communication; valstybinė kolegija / Higher Education Institution, artificial intelligence. She has authored and co-authored Lithuania, a member of Lithuanian Union of Scientists of over 40 scientific publications published in national (LMS), EERA, RENET, LiMA (Lithuanian Marketing and international journals. Association). Her research interests include issues of Vpliv povezave z blagovno znamko in podobe blagovne znamke na prepoznavnost blagovne znamke lito- vskih podjetij digitalnih izdelkov Namen: Prepoznavnost blagovne znamke omogoča večjo prepoznavnost v različnih tržnih razmerah, kar je izred- no pomembno za podjetja, ki prodajajo digitalne izdelke ali ponujajo digitalne storitve. Namen tega prispevka je raziskati učinek povezovanja z blagovno znamko in podobe blagovne znamke na prepoznavnost blagovne znamke litovskih podjetij, ki prodajajo digitalne izdelke. Ti dejavniki so pomembni in vplivajo na prepoznavnost blagovne znamke, vendar njihov vpliv na prepoznavnost blagovne znamke, še posebej v primeru digitalnih izdelkov, doslej ni bil podrobneje raziskovan. Metode: Vprašalnik je bil izveden na vzorcu 403 potrošnikov, ki kupujejo in uporabljajo digitalne izdelke v Litvi. Izde- lan je bil model večkratne linearne regresije (MLR), ki je pokazal statistično pomemben neposreden pozitiven vpliv povezovanja z blagovno znamko in imidža blagovne znamke na prepoznavnost blagovne znamke podjetij digitalnih izdelkov. Rezultati: Empirična raziskava je pokazala statistično pomemben neposreden pozitiven vpliv povezovanja z bla- govno znamko in podobe blagovne znamke na prepoznavnost blagovne znamke litovskih podjetij digitalnih izdelkov. Najmočnejši neposreden pozitiven vpliv na prepoznavnost blagovne znamke ima povezovanje z blagovno znamko. Zaključek: Ugotovitve raziskave zagotavljajo dragocen vpogled v ključno vlogo, ki jo imata povezava z blagovno znamko in podoba blagovne znamke pri gradnji prepoznavnosti blagovne znamke za podjetja, ki ponujajo digitalne izdelke. Ključne besede: Prepoznavnost blagovne znamke, Povezava z blagovno znamko, Podoba blagovne znamke, Dig- italni izdelki 400 AUTHOR GUIDELINES / NAVODILA AVTORJEM Manuscripts considered for publication in Organ- Detailed Guidelines for Authors are available at Članek naj bo razčlenjen v oštevilčena poglavja. izacija (organizacija.fov@um.si) are those which: https://sciendo.com/journal/orga - for Authors. Naslovi članka, poglavij in podpoglavij naj bodo • Contain original work – which is not published All the papers will be reviewed by at least two ref- napisani z malimi črkami, da so razvidne kratice. elsewhere in any medium by the authors or erees. Based on the opinions and suggestions of Slike in tabele v elektronski obliki vključite kar v anyone else and is not under consideration for the reviewers, the editors accept the paper, demand besedilo. Besedilu so lahko priložene slike in/ali ta- publication in any other medium. The author(s) minor or major enhancements, or reject the paper. bele na papirju v obliki pripravljeni za preslikavo. V is/are also responsible for any violations of the If major enhancements are required the upgraded copyright regulations. tem primeru naj bo vsaka slika na posebnem listu, paper is reviewed again. oštevilčene naj bodo z arabskimi številkami, v bese- • Are focused on the core aims and scope of the Manuscripts can be submitted via journal web dilu naj bo označeno, kam približno je treba uvrstiti journal: Organizacija is an interdisciplinary site (https://organizacija.fov.um.si). For further peer reviewed journal that seeks both sliko: na tem mestu naj bo številka slike/tabele in information and clarifications contact Organizaci- theoretically and practically oriented research njen podnapis. Slike bomo praviloma pomanjšali ja’s editorial office (organizacija.fov@um.si or papers from the area of organizational science, maja.mesko@um.si). in jih vstavili v članek. Upoštevajte, da morajo biti business information systems and human oznake in besedila na vseh slikah dovolj velika, da resources management. Address of the Editorial office: bodo čitljiva tudi pri velikosti slike, kot bo obja- • Are clearly and correctly written – should University of Maribor, vljena v reviji. Vse slike naj bodo črno-bele z be- contain all essential features of a complete Faculty of Organizational Science scientific paper, should be written in a clear, Kidričeva cesta 55a lim ozadjem; barvnih slik v tiskani verziji revije ne moremo objaviti, barve so vidne le v spletni verziji. easy to understand manner and be readable for 4000 Kranj, Slovenia a wide audience. Phone: +386-4-2374-297 Članki morajo biti pred objavo v Organizaciji lekto- • Are written in English – should be clearly and rirani. Končno verzijo mora lektorirati naravni go- grammatically written, in an easily readable vorec oz. lektor s primerljivim znanjem angleščine. V reviji Organizacija objavljamo znanstvene član- style. Attention to detail of the language will ke, rezultate raziskovalnega dela avtorjev. Predlože- Podrobna navodila avtorjem za pisanje in oblikova- avoid severe misunderstandings which might ni prispevki naj bodo napisani v angleškem jeziku. nje člankov so na https://sciendo.com/journal/orga lead to rejection of the paper. Correct language Imeti morajo strukturo IMRAD, ki je običajna za is the responsibility of the authors. Unless the - for Authors. znanstvena in strokovna besedila. Objavljamo de- author is an English native speaker, the paper la s predmetnega področja revije, ki še niso bila Predložene prispevke pregledata in ocenita najmanj must be proofread by a language editor, English objavljena in niso bila poslana v objavo v kakšni dva recenzenta. Na osnovi mnenj in predlogov re- native speaker drugi reviji ali zborniku. Avtorji so odgovorni za vse cenzentov uredniški odbor ali urednik sprejmejo All parts of the manuscript should be type-writ- morebitne kršitve avtorskih pravic. prispevek, zahtevajo manjše ali večje popravke in ten (font size 12), with margins of 2.5 cm. Pages Besedilo naj bo oblikovano za tiskanje na papirju dopolnitve ali ga zavrnejo. Če urednik oziroma re- should be numbered consecutively throughout the in levo poravnano. Na začetku prispevka, takoj za cenzenti predlagajo večje popravke, se dopolnjeni manuscript. The text should be subdivided into naslovom, naj bo povzetek (izvleček) dolžine naj- numbered chapters. Figures and tables, consecu- prispevek praviloma pošlje v ponovno recenzijo. več 250 besed, ključne besede, v končni – sprejeti tively numbered (Figure 1, Figure 2, ...; Table 1, verziji članka pa na koncu prispevka tudi kratek Članke za objavo lahko predložite preko spletne- Table 2, …) can be included in electronic form in strokovni življenjepis vsakega od avtorjev (do 10 ga mesta https://organizacija.fov.um.si/index.php/ the text. All lettering and figure elements must be vrstic) in letnica rojstva (zaradi vnosa podatkov v organizacija/information/authors. Za nadaljnje large enough to be readable when figure or table knjižnični informacijski sistem COBISS, v reviji has been reduced to fit journal page or column. All informacije in pojasnila se lahko obrnete na ure- letnica ne bo objavljena). Na prvi strani besedila naj figures and tables must be specifically referred in dništvo Organizacije (organizacija.fov@um.si ali bodo napisani le naslov prispevka, imena in (poštni the text. Colour pictures cannot be published in maja.mesko@um.si). in elektronski) naslovi avtorjev članka, po možnosti the printed version of the journal; colours appear tudi telefonska številka enega od avtorjev. Da bi za- Naslov uredništva: only in the internet version. The paper should gotovili anonimnost recenziranja, naj se imena av- start with a cover page with names and mailing Univerza v Mariboru, torjev ne pojavljajo v besedilu prispevka. Na koncu and electronic addresses of the authors. To assure Fakulteta za organizacijske vede članka, za življenjepisi, naj bo slovenski prevod the anonymity of the refereeing procedure the Kidričeva cesta 55a naslova, povzetka in ključnih besed. names of the authors should not appear in the text. 4000 Kranj Tel.: 04-2374-297 Prva slovenska revija za organizacijska in kadrovska raziskovanja in prakso. Revijo sofinancira Javna agencija za raziskovalno dejavnost Republike Slovenije. Ponatis in razmnoževanje deloma ali v celoti brez pisnega dovoljenja nista dovoljena. Izdajatelj: Univerza v Mariboru, Fakulteta za organizacijske vede Kranj, Založba MODERNA ORGANIZACIJA, Kidričeva cesta 55a, KRANJ, telefon: 04 23 74 200, E-pošta: organizaija.fov@um.si. Uredništvo revije: Kidričeva cesta 55a, 4000 Kranj. Letna naročnina: 47 EUR + ddv. Na leto izidejo 4 številke. Tisk: Tiskarna Koštomaj d.o.o. Naklada 130 izvodov. Organizacija is covered by the following services: Cabell’s Directory, CEJSH (The Central European Journal of Social Sciences and Humanities), Celdes, Clarivate Analytics - Emerging Sources Citation Index (ESCI), CNPIEC, Die Elektronische Zeitschriftenbibliothek, DOAJ, EBSCO - TOC Premier, EBSCO Discovery Service, ECONIS, Ergonomics Abstracts, ERIH PLUS, Google Scholar, Inspec, International Abstracts in Operations Research, J-Gate, Microsoft Academic Search, Naviga (Softweco), Primo Central (ExLibris), ProQuest - Advanced Polymers Abstracts, ProQuest - Aluminium Industry Abstracts, ProQuest - Ceramic Abstracts/World Ceramics Abstracts, ProQuest - Composites Industry Abstracts, ProQuest - Computer and Information Systems Abstracts, ProQuest - Corrosion Abstracts, ProQuest - Electronics and Communications Abstracts, ProQuest - Engineered Materials Abstracts, ProQuest - Mechanical & Transportation Engineering Abstracts, ProQuest - METADEX (Metals Abstracts), ProQuest - Sociological Abstracts, ProQuest - Solid State and Superconductivity Abstracts, Research Papers in Economics (RePEc), SCOPUS, Summon (Serials Solutions/ProQuest), TDOne (TDNet), TEMA Technik und Management, WorldCat (OCLC) CONTENTS - 4/2024 Pavel PELECH, Jaroslava DĚDKOVÁ 319 An Examination of Generational Differences in the Sharing Economy: Understanding the Motivations of Idle Asset Owners and Their Impact on Peer-to-Peer Platform Management Eva JEREB, Marko URH 333 The Use of Artificial Intelligence among Students in Higher Education Aykut ARSLAN, Serdar YENER, Fatma KORKMAZ, Murat Sakir EROGUL 346 Multilevel Investigation of Leadership Prototype Perception: Political Behavior in Relation to Effectiveness and Trust Barbara DORIĆ, Dinko PRIMORAC, Mirjana PEJIĆ BACH 363 Government Effectiveness in the Petroleum Sector: Two-step Analysis Combining Linear Regression and Artificial Neural Networks Božidar LENARČIČ, Armand FAGANEL 379 Using the Nominal Group Technique to Design an Appropriate Communication Strategy Among Strategic Procurement and Internal Stakeholders Gita ŠAKYTĖ-STATNICKĖ 392 The Effect of Brand Association and Brand Image on the Brand Awareness of Lithuanian Companies of Digital Products