Systematic Bibliometric Analysis of Entrepreneurial Intention and Behavior Research

Abstract

Entrepreneurship serves as a vital engine of economic development, yet the mechanisms translating entrepreneurial intention into behavior have gradually emerged. This study employs bibliometric analysis of 61 SSCI-indexed articles (2014–2024) using CiteSpace to examine co-authorship networks, co-citation patterns, and research hotspots. Our findings demonstrate that individual-level factors (personality traits, entrepreneurial self-efficacy, and entrepreneurship education) drive both entrepreneurial intention and entrepreneurial behavior. More importantly, environmental factors (university milieu, regional social legitimacy, and national cultural dimensions) moderate the relationship between entrepreneurial intention and behavior. The study also identifies a temporal pattern in the entrepreneurial intention–behavior correlation. These results advance theoretical understanding of the intention–behavior transition and offer practical insights for entrepreneurship education and policy design.

1. Introduction

Entrepreneurship has been studied for decades, and its significance stems from its influence on economic and social growth. Entrepreneurship offers value to society, solves societal issues, and increases personal wealth (Gibb & Hannon, 2006). Entrepreneurship, particularly entrepreneurial intention (EI), has been extensively researched, and there are relevant literature reviews to summarize entrepreneurial intention, and most scholars mainly conduct literature reviews on EI research (e.g., Delanoë-Gueguen & Liñán, 2019; Dolhey, 2019; Liñán & Fayolle, 2015). However, entrepreneurial intention is not the final destination for starting an entrepreneurial business. Following the suggestion of Fayolle and Liñán (2014) and Kautonen et al. (2013) for future research on the entrepreneurial process, much entrepreneurship research has begun to shift downstream in recent years, that is, to entrepreneurial behavior (EB). While scholars widely assume that stronger intentions directly lead to entrepreneurial behavior, empirical evidence paints a more nuanced picture, showing that such a premise is not always the case (Shirokova et al., 2016; Li et al., 2020). Studies reported that the transition from intention to behavior varies significantly, with conversion rates ranging from as low as 20% to as high as 70% (e.g., Shinnar et al., 2018; Van Gelderen et al., 2018; Neneh, 2019b; Mei et al., 2022). A recent meta-analysis by Tsou et al. (2023) further revealed that entrepreneurial intention explains only 17% of the variance in entrepreneurial behavior, highlighting a significant gap between intention formation and behavioral enactment. Importantly, practical insights suggest that this intention–behavior gap is not uniform but rather highly contingent on contextual factors. Emerging scholarship reveals that entrepreneurial behavior manifests through a dynamic interplay of factors across multiple lenses. From a psychological lens, key drivers like resilience, risk tolerance, and proactive motivation empower aspiring entrepreneurs to transform intention into behavior (Van Gelderen et al., 2015; Li et al., 2020; Kong et al., 2020; Mothibi et al., 2024). These personal factors interact with institutional support systems, where vibrant university ecosystems and robust incubator networks provide essential scaffolding for entrepreneurial development (Shirokova et al., 2016). Beyond these immediate supports, the broader environment plays an equally crucial role, regional entrepreneurial culture and supportive policy frameworks create the fertile conditions necessary for ventures to launch and thrive (Kibler et al., 2014; Bogatyreva et al., 2019; L. X. He & Li, 2024).

While existing research has extensively examined the direct and indirect factors shaping entrepreneurial intention and behavior, including moderators of the intention–behavior transition, significant gaps remain in our understanding. These findings highlight the urgent need to synthesize the proliferating yet disconnected empirical evidence on multi-level factors shaping the entrepreneurial intention–behavior relationship (Rohanaraj, 2023). Although intention–behavior studies have surged in recent years, comprehensive literature reviews on intention–behavior studies remain surprisingly scarce. The few available reviews (e.g., Abbasianchavari & Moritz, 2021; Tsou et al., 2023) tend to focus narrowly on specific topics. For instance, Abbasianchavari and Moritz’s (2021) systematic review highlights role models as a pivotal factor, analyzing how different types of role models and exposure contexts influence the intention–behavior relationship. To narrow this gap, this study conducted a literature review on entrepreneurial intention and behavior research. This study retrieved the literature published between 2014 and 2024 from the Social Sciences Citation Index (SSCI) of the Web of Science Core Collection (WoSCC) and used CiteSpace software to explore the co-authorship contribution, co-cited contribution, and research hotspots. In addition, this study summarized the current status with an analytical framework and proposed future research directions. Thus, to narrow the research gaps, the study aims to answer the following research questions:

During the period from 2014 to 2024,

• RQ1. How has entrepreneurial intention and behavior research evolved in volume?
• RQ2. What patterns emerge in co-authorship (authors/institutions/countries) and co-citation (journals/references/authors)?
• RQ3. What factors directly influence entrepreneurial intention and behavior, and what factors serve as moderating roles in the transition from entrepreneurial intention to behavior?
• RQ4. How does the strength of entrepreneurial intention and behavior relationships change across different time intervals?
• RQ5. Which theoretical frameworks dominate current research, and where do gaps persist?
• RQ6. What are future research opportunities?

The novelty of this study is three fold. First, this study explores the current research profile on EI and EB research rather than only focusing on EI research. Second, this study provides a self-developed framework based on existing EI and behavior research findings. Third, based on the current research gaps, this study proposes future research opportunities.

This study contributes to entrepreneurship literature in four aspects. First, this study reveals the annual number focusing on the EI and EB literature of the last decade. Second, this study illustrates the contribution landscape of co-authorship regarding authors, institutions, and countries, as well as co-citation regarding journals, references, and authors. Third, this study reveals the research hotspots based on keyword co-occurrence analysis. Fourth, this study firstly provides an analytical framework for a better understanding of the variables (independent and moderating ones) on EI and EB from individual (e.g., Li et al., 2020; Neneh, 2019a, 2019b) and environmental factors (e.g., Bogatyreva et al., 2019; Kibler et al., 2014; Shirokova et al., 2016; Murad et al., 2024), the characteristics of longitudinal studies (e.g., Baluku et al., 2020; González-López et al., 2021; Kibler et al., 2014; Neneh, 2019b), the correlation between EI and behavior among longitudinal studies over time (e.g., Joensuu-Salo et al., 2020; Shinnar et al., 2018) and the adopted models (e.g., the TPB model (Ataei et al., 2021; Kautonen et al., 2015; Kibler et al., 2014; Rauch & Hulsink, 2015), and the action phase model (e.g., Van Gelderen et al., 2018)) and different theories (e.g., social role theory (Shinnar et al., 2018), regret regulation theory (e.g., Neneh, 2019b), regret regulation theory (e.g., Neneh, 2019b), and social cognitive career theory (e.g., Belchior & Lyons, 2021)).

The study starts with methods for retrieving the 61 articles in EI and behavior research, followed by results and discussion with visualization using CitaSpace and a self-developed framework. In the next section, this study presents a conclusion with summarized findings, provides theoretical implications by proposing further research opportunities, and provides practical implications for policymakers and educators.

2. Methods

2.1. Data Source and Search Procedure

On 25 April 2025, we conducted an online literature search using the Social Sciences Citation Index (SSCI) of the Web of Science Core Collection (WoSCC). On the same day, we searched for and saved the relevant information. To avoid bias due to daily database updates, we searched the articles using the following query: ((TI = (entrepreneurial intention and action)) OR TI = (entrepreneurial intention and behavior)) AND PY = (2014–2024). The whole procedure of searching is shown in Figure 1. In brief, we only included the type of results “articles” and excluded other types of results (e.g., early access, review articles, and editorial materials). Finally, 190 papers were obtained. After excluding 22 non-article papers (proceeding paper, early access, meeting abstract, book review, review article, editorial material), 168 were initially identified. After excluding 4 non-English articles, 164 articles were identified. After carefully reviewing the abstract of each article, 103 irrelevant articles (only focusing on EI or non-entrepreneurship topics) were excluded and 61 articles were finally identified. For further analysis, this study concentrated solely on the field of entrepreneurial intention and behavior.

2.2. Research Method

This research used CiteSpace (version 6.2.R2) for network analysis and visualization. The software was developed by Dr. Chen from Drexel University in the United States. This software is based on bibliometrics to visually analyze specific research fields and more intuitively identify the contribution, hotspots, and trends of particular research fields (Chen, 2006). In this study, we started with the annual publications in EI and behavior research. Using CiteSpace, this study explores the main contribution of authors, institutions, countries as well as co-cited journals, references, and authors. This study also provides research hotspots with keyword co-occurrence and clustering analysis. Finally, this study provides a self-developed analytical framework based on the literature findings and a summary of current research and ends with proposing further research directions.

3. Results and Discussion

3.1. Publication Trends

Based on the keywords, results’ type and publishing years, and language, 61 publications on entrepreneurial intention and behavior matched the retrieval criteria. To ensure the possible results included publications from a specific time period, we set the publishing time to “2014–2024”. However, the search results indicated that the first publication appeared in 2014, and the number of articles reached 11 in 2024, with the most significant number of publications in the past three years (See Figure 2). In other words, based on our selected publishing time, the number of publications in EI and behavior field has experienced two stages: steady growth and mass emergence. In the first stage (2014–2017), the publishing number had a slow increase. From 2018 to 2024, a relatively large number of relevant research papers emerged, which is expected to grow steadily in 2025.

3.2. The Co-Authorship Analysis

The co-authorship analysis includes the author, institution and country contribution network maps.

3.2.1. Analysis of Author Contribution

Analysis of the author contribution network map helps us to understand the contributing authors and their collaborators in EI and EB research. We finally obtained an author contribution network map in EI and EB research (see Figure 3). Each node in the figure represents an author, the node’s size represents the author’s published articles, the connection between nodes reflects the cooperative relationship between different authors, and various colors indicate the year the author published the article. Specifically, the number of nodes (N) is 71, the number of connections (E) is 49, and the network density is 0.0197. Results include a total of 71 authors and 49 connections between authors. Moreover, the number of connections is lower than that of nodes, and the density is also relatively lower. The findings suggest that while influential scholars have emerged in this research domain, their collaborative networks remain relatively fragmented, with limited cross-author communication and cooperation evident in the publication patterns. In addition, there is close cooperation among several authors headed by Fink, Matthias; Kautonen, Teemu; Duong, Cong Doanh. Duong, Cong Doanh’s publications were mainly published in recent years (2023 and 2024), and Fink, Matthias is an early contributing author (2014 and 2015). We revealed the top 10 authors with the number of publications (see

Table A1. The top 10 authors with publication contributions.

Rank	Count	Centrality	Year	Authors
1	3	0	2014	Fink, Matthias
2	3	0	2014	Kautonen, Teemu
3	3	0	2023	Duong, Cong Doanh
4	2	0	2021	Ashraf, Sheikh Farhan
5	2	0	2019	Shirokova, Galina
6	2	0	2022	Botha, Melodi
7	1	0	2019	Anisimova, Tatiana
8	1	0	2014	Namatovu-dawa, Rebecca
9	1	0	2017	Giones, Ferran
10	1	0	2024	Bashar, Shafiul

), where Fink, Matthias (three publications) was ranked first, followed by Kautonen, Teemu (three publications) and Duong, Cong Doanh (three publications).

3.2.2. Analysis of Institution Contribution

Analysis of the institution contribution network map helps us to grasp the core research institutions in EI and EB research. We finally obtained an institution contribution network map in EI and EB research (see Figure 4). Each node in the figure represents a research institution. The node size represents the number of published articles by the institution, the connections between nodes reflect the cooperative relationship between different institutions, and various colors indicate the year the institution published the article. Specifically, the number of nodes (N) is 73, the number of connections (E) is 58, and the network density is 0.0221. The results include a total of 73 institutions and 58 links between institutions. Moreover, the number of connections is lower than that of the nodes, and the density is relatively low, indicating limited collaboration between contributing institutions. In addition, the collaboration between several institutions headed by Aalto University and Anglia Ruskin University is relatively close and has formed a specific scale. In detail, National Economics University—Vietnam and Shanghai University of Finance & Economics had more publications in recent years (2023 and 2024). Also, Saint Petersburg State University had contributions from 2016 to 2019. However, Aalto University, Aalto University, and Johannes Kepler University Linz contributed to the field in the early years (2014–2016). Publications from the top 10 institutions (see

Table A2. The top 10 institutions with publication contributions.

Ranking	Count	Centrality	Year of First Release	Institutions
1	3	0.01	2014	Anglia Ruskin University
2	3	0.01	2014	Aalto University
3	3	0	2014	Johannes Kepler University Linz
4	3	0	2014	Makerere University
5	2	0	2019	Saint Petersburg State University
6	2	0	2023	National Economics University—Vietnam
7	2	0	2024	Shanghai University of Finance & Economics
8	2	0	2014	Philipps University Marburg
9	2	0	2024	Nankai University
10	2	0	2024	Nanjing University of Finance & Economics

) account for more than 39% of total publications. In detail, Anglia Ruskin University (3), Aalto University (3), Johannes Kepler University Linz (3) and Makerere University (3) ranked first, followed by Saint Petersburg State University (2), National Economics University—Vietnam (2), and Shanghai University of Finance & Economics (2).

3.2.3. Analysis of Country Contribution

Analysis of the country contribution network map helps us to grasp the core research countries in EI and EB research. We finally obtained a country contribution network map in EI and EB research (see Figure 5). Each node in the graph represents a research country, the node’s size represents the number of published articles by the country, the connections between nodes reflect the cooperative relationship between countries, and different colors indicate the year in which the article was published in that country. Specifically, the number of nodes (N) is 38, the number of connections (E) is 47, and the network density is 0.0669. The results include a total of 38 countries and 47 links between countries. The number of connections is higher than that of countries, which indicates that countries that have made significant contributions in this field have close cooperation and a relatively high degree of communication. In addition, the collaboration between countries headed by China, USA, and England is relatively close and has formed a particular scale. In detail, the USA made a continuous contribution from 2014 to 2024. Similarly, Finland contributed to this field from 2014 to 2020. Although China started later in 2020, the country had the most publications (See Figure 5). According to the top 10 countries participating in the study of entrepreneurial intention and behavior (See

Table A3. The top 10 countries with publication contributions.

Ranking	Count	Centrality	Year of First Release	Countries
1	19	0.47	2020	China
2	6	0	2019	South Africa
3	6	0.74	2014	England
4	6	0	2019	Pakistan
5	5	0.32	2017	Spain
6	5	0.68	2014	USA
7	4	0	2014	Finland
8	3	0	2023	Vietnam
9	3	0.17	2014	Austria
10	3	0.09	2018	Portugal

), China contributed the most publications (16), followed by the South Africa (6), England (6), and Pakistan (6).

3.3. Co-Citation Analysis

3.3.1. Analysis of Co-Cited Journals

In bibliometric analysis, journal co-citation indicates that two journals are co-cited in at least one article (McCain, 1991). Analysis of co-cited journals provides insight into leading journals important to EI and EB research. We finally obtained a journal co-citation network map in EI and EB research (See Figure 6). Each node in the figure represents a cited journal, the node’s size represents the number of times the journal has been cited, the links between nodes reflect co-citations between different journals, and different colors indicate the year in which the journal was co-cited. Specifically, the number of nodes (N) is 68, the number of connections (E) is 184, and the network density is 0.0808. Results included a total of 68 journals and 183 co-citations between journals. Moreover, the number of connections is higher than that of nodes, and the density is also relatively high, indicating that the journals cited in this field are concentrated. In addition, several journals headed by Entrepreneurship Theory and Practice (IF 2023 = 7.8) and Journal of Business Venturing (IF 2023 = 7.7) are more critical. According to the top 10 co-cited journals (see

Table A4. The top 10 co-cited journals.

Ranking	Freq	Centrality	Year	Cited Journals	IF in 2023
1	56	0.48	2015	Entrepreneurship Theory and Practice	7.8
2	56	0.11	2014	Journal of Business Venturing	7.7
3	48	0.17	2015	Organizational Behavior and Human Decision Processes	3.4
4	44	0.09	2017	Journal of Business Research	10.5
5	41	0.25	2017	International Entrepreneurship and Management Journal	6.2
6	35	0.65	2014	Academy of Management Review	19.3
7	35	0.01	2014	Small Business Economics	6.2
8	33	0.05	2014	Journal of Small Business Management	5.3
9	30	0.1	2017	International Journal of Entrepreneurial Behavior & Research	4.5
10	25	0.34	2015	Entrepreneurship & Regional Development	3.3

), Entrepreneurship Theory and Practice (56) and Journal of Business Venturing (56) ranked first, followed by Organizational Behavior and Human Decision Processes (48) and Journal of Business Research (44).

3.3.2. Analysis of Co-Cited References

In bibliometric analysis, reference co-citation is established when two distinct publications are jointly referenced in the citation list of one or more subsequent documents (Small, 1973). Analysis of co-cited references helps us to understand the primary literature critical in EI and EB research. We finally obtained a reference co-citation network map in EI and EB research (See Figure 7). Each node in the figure represents a cited article, the size of the node represents the number of times the article has been cited, the links between nodes reflect the co-citations between different articles, and different colors indicate the year in which the article was co-cited. Specifically, the number of nodes (N) is 220, the number of connections (E) is 655, and the network density is 0.0272. Results included a total of 220 papers and 655 co-citations between papers. Moreover, the number of connections is higher than that of nodes, and the density is also relatively high, indicating that the literature that has been co-cited in this field is relatively concentrated. In addition, the article of Schlaegel and Koenig (2014) was more influential in the early years because the work was frequently co-cited from 2014 to 2017, and that of Gieure et al. (2020) was widely co-cited in recent years (2020–2024). According to the top 10 co-cited articles (see

Table A5. The top 10 co-cited references.

Ranking	Counts	Centrality	Year	Cited References
1	11	0.71	2020	Gieure et al. (2020)
2	11	0.27	2014	Schlaegel and Koenig (2014)
3	10	0	2015	Kautonen et al. (2015)
4	10	0.05	2015	Liñán and Fayolle (2015)
5	9	0.19	2020	Meoli et al. (2020)
6	9	0.05	2016	Shirokova et al. (2016)
7	9	0.17	2019	Neneh (2019b)
8	8	0.04	2015	Van Gelderen et al. (2015)
9	7	0.09	2019	Bogatyreva et al. (2019)
10	7	0.05	2014	Fayolle and Liñán (2014)

), Gieure et al. (2020) and Schlaegel and Koenig (2014) ranked first (11), followed by Kautonen et al. (2015) (10) and Liñán and Fayolle (2015) (10).

3.3.3. Analysis of Co-Cited Authors

In bibliometric analysis, author co-citation refers to the frequency with which two scholars’ works are jointly referenced by subsequent publications (Y. He & Hui, 2002). Analysis of co-cited authors can help us to understand that the leading authors are influential in EI and EB research. We finally obtained a co-cited author network map in EI and EB research (See Figure 8). Each node in the figure represents a cited document, the size of the node represents the number of times the author has been cited, the links between the nodes reflect the co-citations between different authors, and the different colors indicate the year the author was co-cited. Specifically, the number of nodes (N) is 111, the number of connections (E) is 244, and the network density is 0.04. The results included a total of 111 papers and 244 co-citations between authors. Moreover, the number of connections is higher than that of nodes, and the density is also relatively high, indicating that the co-authors’ literature is relatively concentrated in this field. In addition, Ajzen I and Krueger NF are influential because they are co-cited from 2014 to 2024 and have higher co-citations. According to the top 10 co-cited authors (See

Table A6. The top 10 co-cited authors.

Rank	Count	Centrality	Year	Cited Authors
1	51	0.63	2014	Ajzen I
2	42	0.05	2016	Liñán F
3	38	0.05	2015	Kautonen T
4	33	0.23	2015	Krueger NF
5	22	0.47	2016	Fayolle A
6	22	0	2019	Van Gelderenm
7	18	0.04	2019	Shirokova G
8	16	0.42	2015	Bird B
9	15	0	2020	Fornell C
10	15	0.04	2017	Schlaegel C

), Ajzen I (51 co-citations) was ranked first among the top 10 co-cited authors, followed by Linan F (42 co-citations), Kautonen T (38 co-citations), and Krueger NF (33 co-citations).

3.4. Analysis of Research Hotspots

3.4.1. Keyword Co-Occurrence Analysis

The keyword is the vocabulary the author refined to summarize the article’s subject. It is the author’s academic thought, research topic, and content for specific research. We can obtain a specific area’s research topics and hotspots by analyzing keywords. In the keyword contribution network pictures constructed by CiteSpace software, each node represents a keyword, and the size of each node represents the frequency of the keyword’s appearance. Centrality is used to quantify the importance of a keyword’s position in the network. After eliminating the topic keyword “entrepreneurial intention” and unified the same meaning keywords, we finally obtained the keyword co-occurrence network map (see Figure 9). Specifically, the figure includes a total of N = 202 nodes, E = 864 connections, and the network density = 0.0426. The keyword co-occurrence analysis revealed several prominent research themes in this domain. The high-frequency keywords included “theory of planned behavior,” reflecting the Theory of Planned Behavior’s dominant role as the prevailing theoretical framework. Beyond the model, the analysis identified several key topics frequently examined in entrepreneurial intention and behavior research, such as “self-efficacy,” “entrepreneurship education,” “university students,” “gender,” and “link.” Notably, the transition from intention to action was represented by keywords “link”.

3.4.2. The Analysis of Research Hotspots by an Analytical Framework

Building upon the preceding discussion, we systematically examine research hotspots through three critical dimensions: (1) influential factors, (2) the intention–behavior transition, and (3) predominant theoretical models.

Entrepreneurial intention, conceptualized as the conscious commitment to establish a new venture (Krueger, 1993), serves as fundamental predictors of entrepreneurial behavior. However, empirical evidence reveals a significant intention–behavior gap, where many individuals with strong entrepreneurial intentions ultimately delay or abandon venture creation (Kautonen et al., 2015). This discrepancy stems from various uncertainty factors, including fluctuating personal preferences and emergent situational constraints. Our comprehensive literature review identifies key research patterns: (1) direct determinants shaping either entrepreneurial intention or entrepreneurial behavior; and (2) the moderating effects on the EI-EB relationship. These relationships are visually synthesized in Figure 10, which presents an integrated framework of the EI-EB transition process.

The Direct Effects on Entrepreneurial Intention and Behavior

The formation of entrepreneurial intentions and subsequent behavioral enactment are governed by distinct yet overlapping mechanisms. In the intention formation phase, individuals’ entrepreneurial aspirations emerge through an interplay of (a) dispositional characteristics (autonomy orientation), (b) cognitive factors (entrepreneurial self-efficacy, entrepreneurial competency, cognitive style, and entrepreneurship education), and (c) contextual factors (institutional support) (Delanoë-Gueguen & Liñán, 2019; Murad et al., 2024; Zhuang & Sun, 2023a, 2023b). Crucially, when transitioning to actual behavior, while core antecedents like cognitive style, entrepreneurial self-efficacy, and entrepreneurship education maintain their influence, three additional determinants become pivotal: (1) role models, (2) managing the whole process, and (3) affective impediments, particularly failure aversion, that collectively influence the entrepreneurial behavior (Delanoë-Gueguen & Liñán, 2019; Kong et al., 2020; Zhuang & Sun, 2024).

Moderating Effects Between Entrepreneurial Intention and Behavior

Individual factors. Research identifies several individual-level moderators in the EI-EB relationship, including fear of failure, role models (Kong et al., 2020), action-related emotions (aversion, fear, doubt), self-control (Van Gelderen et al., 2015), and gender (Shinnar et al., 2018). Kong et al.’s (2020) cross-sectional study revealed fear of failure negatively moderates EI-EB while role models positively influence behavior. Van Gelderen et al.’s (2015) longitudinal study demonstrated self-control’s positive moderating effect, whereas Shinnar et al. (2018) found gender moderates EI’s conversion to start-ups three years after graduation. Job security similarly emerges as a negative moderator (Delanoë-Gueguen & Liñán, 2019), with gender differences explained by societal roles and family responsibilities that disproportionately affect women (Hsu et al., 2016; Justo et al., 2015; Zhao et al., 2005).

Key personality moderators include proactive personality (Li et al., 2020; Neneh, 2019a), trait competitiveness, entrepreneurial alertness (Neneh, 2019a), anticipated regret (Neneh, 2019b), and risk-taking propensity (Mothibi et al., 2024). Studies demonstrate proactive personality’s positive moderating role across cultural contexts (China: Li et al., 2020; South Africa: Neneh, 2019b), while anticipated regret similarly strengthens EI-EB conversion (Neneh, 2019b). However, risk-taking propensity weakens this relationship (Mothibi et al., 2024). These traits operate through distinct mechanisms: proactive individuals actively shape their environment (Delle & Amadu, 2016), competitive individuals pursue goals to “win” (Fuller et al., 2018), and regret-averse individuals act to avoid negative future emotions (Pieters & Zeelenberg, 2007).

Environmental factors. Environmental factors significantly influence the transition from entrepreneurial intention to entrepreneurial behavior, including regional social legitimacy (Kibler et al., 2014), culture (Bogatyreva et al., 2019), university milieu (Shirokova et al., 2016), environmental uncertainty (L. X. He & Li, 2024), and entrepreneurship opportunity. Regional social legitimacy refers to the extent entrepreneurial activities are perceived as socially “desirable, proper, or appropriate” (Suchman, 1995). Kibler et al. (2014) found that social legitimacy strengthens the EI-EB relationship in underdeveloped regions, where institutional support is crucial. However, this effect diminishes in high-GRP (gross regional product) areas, as stable employment opportunities increase the opportunity cost of entrepreneurship.

Cultural dimensions shape entrepreneurial transitions by influencing risk perceptions and societal expectations (Hechavarría, 2016). Bogatyreva et al. (2019), analyzing GUESS data (N = 1435 students), found that individualism, long-term orientation, indulgence, power distance, and uncertainty avoidance negatively moderate the EI-EB link, whereas masculinity (competitive, achievement-oriented cultures) strengthens it.

University environment also plays a critical role. Shirokova et al. (2016) demonstrated that supportive university ecosystems (e.g., incubators, mentorship) enhance EI-EB transition. Entrepreneurial motivation further moderates this transition. Santos et al. (2021) distinguished between necessity-driven (extrinsic motivation due to job scarcity) and opportunity-driven (intrinsic motivation to exploit market gaps) entrepreneurship, finding that opportunity-driven individuals exhibit a stronger EI-EB relationship. Lastly, environmental uncertainty weakens the EI-EB relationship (L. X. He & Li, 2024), as unstable economic conditions heighten perceived risks.

Longitudinal Surveys with Different Data Waves and the Correlation of Transition

Scholars have increasingly called for moving beyond intention-centric models to better understand the temporal translation process from entrepreneurial intention to behavior (Shirokova et al., 2016). Our systematic review reveals that this intention–behavior transition is not merely moderated by contextual factors but is fundamentally temporally embedded, a dimension critically reflected in longitudinal research designs. The extant literature demonstrates significant heterogeneity in temporal sampling approaches:

(1)

Short-cycle studies (3–6 months):

• Van Gelderen et al. (2018): Swedish population (N = 422), 6-month interval;
• Neneh (2019b): South African students (N = 308 → 277), 6-month interval;
• Baluku et al. (2020): Student cohort (N = 149), 4-month interval.

(2)

Annual-cycle studies:

• Kibler et al. (2014): Finnish (N = 732) & Austrian (N = 252) samples, 1-year interval;
• Van Gelderen et al. (2015): Two-wave survey (2011–2012; 1-year interval), Finnish adults (N = 703).

(3)

Multi-year longitudinal studies:

• Liñán and Rodríguez-Cohard (2015): Three waves spanning 2004–2007;
• Delanoë-Gueguen and Liñán (2019): Two-wave study with 5-year interval (N = 155);
• Santos et al. (2021): Decadal tracking (2010–2019, N = 593).

Empirical evidence indicates that the predictive validity of entrepreneurial intention for subsequent behavior is temporally sensitive. While entrepreneurial intention typically explains 30–45% of entrepreneurial behavior variance (Kautonen et al., 2013; Shirokova et al., 2016; Van Gelderen et al., 2015), longitudinal studies reveal the critical nuance:

(4)

Temporal Decay in Predictive Strength (notes: *: p-value < 0.05; **: p-value < 0.01; ***: p-value < 0.001.)

• The EI-EB correlation remains significant but attenuates over time (Shinnar et al., 2018):
  Short-term: 0.66 ** (1 semester) → 0.97 ** (6 months);
  Long-term: 0.78 * (3 years).
• Similar decay observed by Joensuu-Salo et al. (2020):
  0.727 *** (1–3 years after graduation) vs. 0.605 * (6–8 years after graduation).

These findings collectively demonstrate that while the EI-EB relationship remains statistically significant across time, its effect strength follows a clear temporal decay pattern. The evidence suggests entrepreneurial intention serves as a necessary but increasingly insufficient predictor of eventual behavior, with predictive capacity diminishing as the intention–behavior gap widens. This temporal contingency raises critical theoretical questions about whether entrepreneurial intention alone constitutes a sufficient condition for entrepreneurial behavior formation, warranting further investigation into (1) the minimum intention thresholds required for behavior, and (2) the compensatory mechanisms that sustain behavioral enactment over extended periods.

The Models and Theories Adopted in Entrepreneurial Intention and Behavior Research

Our literature review found some theories and models to explain the EI-EB relationship. Ajzen (1991) developed the theory of planned behavior (TPB) theory, which is based on the theory of reasoned (TRA) by Fishbein and Ajzen (1973). The theory of the planned behavior model was widely used in previous empirical studies (e.g., Ataei et al., 2021; Kautonen et al., 2015; Kibler et al., 2014; Rauch & Hulsink, 2015; Zhuang et al., 2022). TPB has three antecedents: attitude toward entrepreneurship, subjective norms, and perceived behavioral control. Attitude toward entrepreneurship refers to individual perception of whether starting a venture is a favorable career choice. Subjective norms refer to individual perceptions of whether significant others support their behaviors of starting a venture. Perceived behavioral control refers to the individual perception of the ease of starting a venture. The second theory is the social cognitive career theory model. Based on the self-efficacy theory (Bandura, 1977) and social cognitive theory, Lent and Brown (2017) proposed the social cognitive career theory, which includes three antecedents: self-efficacy, outcomes expectations, and goal-directed activities. The theory emphasizes the impacts of personal input and environmental effects on personal, intentional action.

The Rubicon model of action phases has been used in few studies to characterize the phase of intention execution (see Figure 10), which begins with the emergence of an entrepreneurial intention and ends with the launch of a new enterprise (González-López et al., 2021; Van Gelderen et al., 2015, 2018). There are four phases in the Rubicon model of the action phase: the pre-decision phase, the action phase, and the post-action phase. In the first stage, would-be business owners evaluate the project’s desirability and feasibility and establish their preferences to start a venture. In the next two stages, known as the pre-actional and actional phases, entrepreneurs engage in gestational behavior to put into action their start-up preparation and establish their business. In the post-action phase, entrepreneurs evaluate the current status of the company and try to envision its future direction.

4. Conclusions and Implications

This study began with searching the Web of Science for articles on entrepreneurial intention and behavior published between 2014 and 2024. This study makes the key contributions to the literature:

First, this study revealed the annual number and trend of publications focusing on EI and EB. Based on 61 articles retrieved from the SSCI database of WoSCC, this study found that research in this field has grown steadily since 2014 and has expanded significantly over the past three years.

Second, this study illustrated the contribution landscape of co-authorship regarding authors, institutions, and countries as well as co-citation regarding journals, references, and authors. The analysis of co-authorship by authors, institutions and countries revealed that there are few cooperations among different authors (e.g., Fink, Matthias, Kautonen, Teemu, and Duong, Cong Doanh), institutions (e.g., Aalto University and Anglia Ruskin University) and countries (e.g., China, USA, and England). In addition, the analysis of co-citation analysis revealed the important journals (e.g., Entrepreneurship Theory and Practice (IF 2023 = 7.8) and Journal of Business Venturing (IF 2023 = 7.7)), references (e.g., Gieure et al., 2020; Schlaegel & Koenig, 2014) and authors (e.g., Ajzen I and Krueger NF).

This study develops a framework for better understanding the effects of independent variables (direct and indirect) and moderating variables on entrepreneurial intention and behavior by synthesizing existing academic research from 2014 to 2024. Among longitudinal studies, most adopted two-wave data collection, with intervals typically within one year (e.g., Baluku et al., 2020; González-López et al., 2021; Kibler et al., 2014; Neneh, 2019b). For studies with three waves, the correlation between entrepreneurial intention and behavior remained stable over months or even years but weakened over time (e.g., Joensuu-Salo et al., 2020; Shinnar et al., 2018).

Regarding the model adoption, the TPB model was widely used in the literature (e.g., Ataei et al., 2021; Kautonen et al., 2015; Kibler et al., 2014; Rauch & Hulsink, 2015; Mothibi et al., 2024; Murad et al., 2024), and some used the action phase model (e.g., Van Gelderen et al., 2018; L. X. He & Li, 2024). There are some other approaches from different theories such as social role theory (e.g., Shinnar et al., 2018), regret regulation theory (e.g., Neneh, 2019b), and social cognitive career theory (e.g., Belchior & Lyons, 2021; Murad et al., 2024).

4.1. Theoretical Implications

This study offers important theoretical implications by identifying key directions for future research. Existing literature suggests that individual factors tend to exert a direct influence on EI and EB. While existing literature has predominantly emphasized the direct effects of individual-level factors (e.g., personality traits, entrepreneurial self-efficacy) on EI and EB, our comprehensive analysis reveals that the EI-EB transition is more accurately conceptualized as a dynamic, multilevel process shaped by the complex interplay of (1) micro-level psychological factors (e.g., cognitive biases, risk propensity); (2) meso-level institutional factors (e.g., university support systems, incubator networks); and (3) macro-level contextual factors (e.g., regional social legitimacy, national cultural dimensions). The time-varying strength of the EI-EB relationship also underscores the necessity of systematically integrating temporal dimensions into theoretical models of entrepreneurial intention and behavior.

This study further identifies three theoretically grounded priority areas for future EI-EB research. First, much of the existing research relies on student samples, raising concerns about the generalizability of findings. To address the student sample bias, future work should compare different demographic groups, such as students versus working professionals, to examine potential differences in opportunity recognition, entrepreneurial skills, and perceptions of the business environment. Second, several underexplored variables warrant deeper investigation. For instance, social capital plays a critical role in facilitating venture creation, yet its impact remains insufficiently examined. Age may also moderate the EI-EB relationship, as younger individuals are typically more inclined toward entrepreneurship than older individuals. This difference may stem from the delayed financial returns associated with starting a business, whereas wage labor provides immediate income. Future research should explore these dynamics to enhance our understanding of entrepreneurial pathways across diverse populations. Third, the majority of existing studies have adopted the Theory of Planned Behavior. Therefore, it is recommended that several other theoretical frameworks be investigated. Many other models of entrepreneurial intention exist as well, such as the entrepreneurial event model (Shapero & Sokol, 1982), which posits that entrepreneurial intention is influenced by a person’s inclination to act as well as their views on the venture’s desirability and feasibility. Furthermore, there are different types of entrepreneurs based on the four phases of the Rubicon model. For those aspiring entrepreneurs who are still in the motivational phase, entrepreneurial intention is a strong and favorable predictor of eventual start-up status. For those nascent entrepreneurs on pre-actional and actional phases, they have high engagement in entrepreneurial activities. For those new entrepreneurs who have successfully registered their companies, they should think about the current venture’s operation and future growth. It is necessary to identify the entrepreneurs in various business phases since venture formation is a long process marked by distinct activities and different stages. Then, we suggested that further research can explore the impacts on aspiring and nascent entrepreneurs by considering the Rubicon model as their conceptual model.

4.2. Practical Implications

This study offers practical implications by demonstrating how personal and environmental factors collectively influence the translation of entrepreneurial intention into entrepreneurial behavior, suggesting that strengthening personality traits and fostering opportunity-driven ecosystems with supportive cultures can effectively nurture this transition. These integrated findings yield important policy implications, suggesting that regionally tailored entrepreneurial policies accounting for multilevel contextual factors and psychological mechanisms can more effectively facilitate the intention–behavior transition. For educators, these results underscore the need to design intervention programs that specifically target psychological barriers, particularly self-control, anticipated regret, and fear of failure, which frequently impede entrepreneurial behavior.

4.3. Limitations and Future Research

This study has several limitations that warrant acknowledgment. First, the analysis relied exclusively on the Social Sciences Citation Index of Web of Science Core Collection, which may have omitted relevant studies indexed in Scopus, Google Scholar, or regional databases, potentially limiting the comprehensiveness of the findings. Second, the inclusion of only English-language publications could introduce language bias, underrepresenting valuable non-English research on EI-EB dynamics. Third, while bibliometric methods effectively map citation networks and thematic trends, they do not assess the qualitative depth or methodological rigor of individual studies. Additionally, reliance on specific keywords (e.g., “entrepreneurial intention and behavior”) may have excluded relevant work using alternative terminology (e.g., “venture creation propensity”).

Building on these limitations and given the nascent stage of EI-EB relationship research with limited empirical evidence, unexamined boundary conditions regarding sample heterogeneity, and inadequate longitudinal validation, we recommend three critical directions for advancing the framework by including updated research: (1) adopting more representative samples to enhance generalizability, (2) conducting competing model analyses to establish theoretical robustness, and (3) implementing multi-wave research designs to elucidate temporal dynamics. These methodological advancements would substantiate the proposed framework while addressing current limitations in scope and predictive validity.