Cross-national Dataset on Psychological, Cognitive, and Contextual Factors of Entrepreneurial Behavior in Business Students

Theoretical background

Over the past fifteen years, research on University Entrepreneurial Behavior (UEB) in Latin America has undergone significant development, driven by regional efforts to promote innovation-based economies and to recognize the role of universities as key agents in cultivating entrepreneurial talent. In this context, the present study analyzes the psychological, cognitive, and contextual factors that influence the entrepreneurial behavior of business students, drawing on regional evidence and theoretical frameworks that explain the determinants of entrepreneurial intention and action.

Regarding psychological factors, the literature on UEB in Latin America has been predominantly grounded in Ajzen’s (1991) Theory of Planned Behavior (TPB), which highlights the role of attitudes, subjective norms, and perceived behavioral control in shaping entrepreneurial intention. Comparative quantitative studies, such as those by López et al. (2021) and López and Álvarez (2019), based on data from the Global University Entrepreneurial Spirit Students’ Survey (GUESSS), confirm that attitudes toward entrepreneurship and perceived control are robust predictors of entrepreneurial intention across different Latin American contexts. Complementarily, recent research has incorporated competency-based approaches (Fernández-Pérez et al., 2019), recognizing that entrepreneurship education not only influences attitudes but also develops specific competencies—such as innovation, proactivity, leadership, and self-efficacy—that mediate the relationship between educational experiences and entrepreneurial intentions.

The competency-based approach proposed by Fernández-Pérez et al. (2019) extends Ajzen’s (1991) cognitive model by including socio-emotional competencies, such as self-awareness, empathy, and social skills. Self-awareness refers to the ability to recognize one’s own emotions and their effects on actions, awareness of strengths and limitations, and having strong confidence in one’s self-worth and abilities. Empathy is the ability to understand others’ feelings and viewpoints, actively take an interest in what they care about, acknowledge the needs of others to progress, and anticipate and recognize the needs of others. Social skills include the ability to use effective means of persuasion, be a good listener, convey a compelling message, negotiate and resolve disagreements, and inspire, lead, and direct change. Although these competencies are not directly related to entrepreneurial intention, they are associated with the cognitive variables of the TPB model, especially attitude toward entrepreneurship and self-efficacy. These results are consistent with previous approaches in Latin America that have included socio-emotional leadership skills and locus of control as antecedents of entrepreneurial intention in university students (Contreras et al., 2017).

Subjective well-being is another psychological factor linked to entrepreneurship. High levels of psychological well-being provide individuals with the emotional and mental strength to successfully manage the rigorous demands and stress inherent to entrepreneurship (Mabhena & Ncube, 2024). Moreover, an empirical study found that subjective well-being’s effect on entrepreneurial intent did not show a direct positive effect, but rather that subjective well-being exerts indirect effects mediated by the subjective norm, a key dimension of the TPB model (Contreras-Barraza et al., 2022).

Moving on to another point, the cognitive factors namely financial knowledge, cognitive load, and flow. First, financial knowledge is a critical factor for entrepreneurial success and sustainability, as evidenced by the high correlation between its absence and startup failure rates. Understanding basic financial principles—such as cash flow management and profitability analysis—allows entrepreneurs to effectively make informed strategic decisions, optimize resource allocation, and negotiate financing. Recent research, such as the study by Abdallah et al. (2025) on SME performance, highlights that the accessibility of finance moderates the positive impact of financial literacy, meaning that the value of the entrepreneur’s knowledge is maximized when mechanisms for accessing capital are available.

Second, cognitive load theory classifies the mental demands on the entrepreneur into three categories: intrinsic load (inherent business complexity), extraneous load (effort wasted on inefficiencies), and Germane load (the productive effort for innovation). Entrepreneurial success depends on minimizing the extraneous load and maximizing the Germane load, thereby ensuring that mental energy is channeled toward the creative solution of core challenges (i.e., intrinsic load) and the development of a sustainable competitive advantage (Parte et al., 2018).

Third, the evidence suggests that a state of flow significantly contributes to the entrepreneur’s subjective well-being and productivity (Bahsri et al., 2023). This vital psychological state, characterized by deep absorption and effortless action, directly influences performance and innovation capacity by unlocking higher levels of creativity and sustained productivity. Research further suggests that flow facilitates more effective decision-making and the generation of novel solutions, as the intense focus allows for the synthesis of complex information without cognitive distraction, ultimately translating into superior operational efficiency and creative output (Patel & Singh, 2024).

Turning now to the third aspect, institutional and cultural factors have also gained relevance in Latin American scholarship. Studies conducted in countries such as Colombia, Peru, Chile, and Ecuador demonstrate that the perception of an entrepreneurial university environment, characterized by role models, institutional support, and a culture oriented toward innovation, significantly predicts entrepreneurial intentions (Soria-Barreto et al., 2017; Leiva et al., 2021).

Despite these advances, important gaps remain in the Latin American UEB literature. Although previous research has advanced the understanding of the factors that influence entrepreneurial intention, an important limitation of the existing literature is that most studies are conducted at the local or national level, which restricts the generalizability of their findings (Romaní et al., 2021). This geographical fragmentation limits the ability to compare how psychological, cognitive, and contextual factors operate across different sociocultural environments. Our study addresses this gap by adopting a regional, cross-national perspective involving five Latin American countries, thereby offering greater external validity and enabling comparative insights. Second, there is insufficient integration among psychological, cognitive, and contextual factors: few studies simultaneously model affective traits (e.g., fear of failure), cognitive processes (e.g., entrepreneurial alertness or implementation intentions), and institutional variables (e.g., university climate or national innovation systems). Third, there are notable geographical imbalances, as research tends to concentrate in countries such as Mexico, Colombia, and Chile, while the realities of Central America, the Caribbean, and parts of the Andes remain underrepresented. Finally, there is a weak connection between competency-based approaches and TPB constructs, since few studies explicitly analyze how competencies relate to attitudes, subjective norms, or perceived control within a unified causal framework.

In light of these gaps, the present study offers a theoretical and empirical contribution that strengthens the regional understanding of university entrepreneurial behavior. By systematically integrating the psychological, cognitive, and contextual dimensions of entrepreneurial behavior within a model grounded in both the TPB and competency-based approaches, the study proposes a more holistic and causally grounded explanation of how business students develop and sustain their entrepreneurial intentions. Furthermore, the study situates this analysis within a comparative Latin American perspective, thereby contributing to the consolidation of a regional body of knowledge capable of guiding educational strategies and public policies aimed at promoting entrepreneurship as a driver of sustainable development and social inclusion.

Participants

The sample for this study consisted of undergraduate students enrolled in business schools accredited by the ACBSP in four Latin American countries: Mexico (n = 170), Peru (n = 100), Ecuador (n = 147), Paraguay (n = 82), and Colombia (n = 42). Diversity was maximized by recruiting from multiple ACBSP-accredited universities in each country. This approach aligns with methodological standards in cross-national research on entrepreneurial education (Fayolle & Gailly, 2015; López et al., 2021).

A voluntary and deterministic sampling method was employed, with inclusion criteria requiring participants to be over 18 years of age and to have completed at least one entrepreneurship-related course at their educational institution. The achieved sample sizes per country exceed the minimum thresholds recommended for cross-cultural comparative analyses (Field, 2013).

Data collection was conducted between May 2024 and May 2025 through coordinated recruitment with program directors, including invitations sent via institutional email, announcements in virtual classrooms, and direct communication in scheduled lectures. The self-administered online survey was delivered through a secure platform and required approximately 15–20 minutes to complete.

A total of 541 valid responses were obtained. The study employed a correlational, cross-sectional, and non-experimental design.

Written informed consent was obtained from all participants before data collection. For the online survey, consent was documented through a mandatory digital acceptance form displayed before the questionnaire, in which participants confirmed their willingness to take part in the study and acknowledged the confidentiality terms. All participants were over 18 years of age; therefore, no parental consent or minor assent procedures were necessary. Participation was entirely voluntary, and the research instrument did not cause any discomfort, physical, or psychological harm. The study was conducted fairly and equitably, ensuring gender balance in the sample. Data confidentiality and anonymity were guaranteed by Decree 1377 of 2013, Law 1581 of 2012, Andean Decision 351 of 1993, and the institutional policy on personal data management.

Measures

Measurement instruments

A set of validated questionnaires was administered, organized into four blocks: psychological, cognitive, contextual, and entrepreneurial. All scales used a 5-point Likert format (1 = Strongly disagree, 5 = Strongly agree), with scores calculated as the mean of the items. The Spanish versions were reviewed through a translation–back translation process and validated by experts. Average Variance Extracted (AVE) and reliability (Cronbach’s α and McDonald’s ω) for the present sample are reported below.

Statistical analysis

Data collection was conducted between May 2024 and May 2025 through recruitment strategies coordinated with program directors from accredited business schools, to ensure diversity in gender, age, and country representation, thereby reducing potential sampling bias.

Measures to minimize common method bias included:

Confidentiality and anonymity were maintained to reduce social desirability bias.

The statistical analysis proceeded in several stages:

All analyses were conducted in JASP (version 0.19.3.0) and Jamovi (version 2.6.44).

Risk

Figure 1 presents the distribution of risk perception by gender and country, revealing noteworthy patterns that may be interpreted through a gender perspective. This raises the question of whether significant differences exist in risk-taking propensity between male and female students across the university contexts studied, and how these differences manifest in each country.

Figure 1. Risk perception by gender and country.

Raincloud plots showing the distribution of risk perception scores among male and female undergraduate business students in Mexico (n = 170), Peru (n = 100), Ecuador (n = 147), Paraguay (n = 82), and Colombia (n = 42). Plots combine half-violin density curves, boxplots (median, interquartile range), and jittered raw data points. Scores measured on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree).

Differences in risk perception by gender and country were analyzed. Although no statistically significant differences were found in most cases (p > .05), group median values showed a non-significant numerical trend toward slightly lower scores among women in countries such as Mexico, Ecuador, and Colombia (medians between 3.0 and 3.5 on a 5-point scale), which may suggest lower risk-taking propensity in these contexts.

Entrepreneurial intention

On the other hand, Figure 2 illustrates the distribution of entrepreneurial intention across the countries studied. The results suggest that having prior entrepreneurial experience does not guarantee a higher level of entrepreneurial intention. This may indicate that experience alone does not significantly differentiate the development of intention within this sample.

Figure 2. Entrepreneurial intention by country.

Raincloud plots showing the distribution of entrepreneurial intention across the five countries. Colors indicate countries; plots combine density, boxplots, and raw points. Scores on a 5-point Likert scale.

Peru and Paraguay stood out as the countries with the highest levels of entrepreneurial intention among students, which may be attributed to sociocultural and institutional factors that foster proactive attitudes toward entrepreneurship. In contrast, students from Mexico and Ecuador showed a more dispersed distribution with a higher presence of low scores, suggesting a weaker consolidation of entrepreneurial intention in these contexts. Colombia, meanwhile, exhibited a high and homogeneous median, although without the extreme values observed in Peru. These differences may be related to the national entrepreneurial climate, perceived opportunities, or the type of training received.

Entrepreneurial attitude

Figure 3 reflects the trend that cumulative exposure to content, practical experiences, and entrepreneurial learning environments positively impacts entrepreneurial attitude—students in higher semesters reported greater levels of entrepreneurial attitude.

Figure 3. Entrepreneurial attitude by academic semester and country.

Raincloud plots showing entrepreneurial attitude levels in relation to academic progress. Data grouped by country. Scores on a 5-point Likert scale.

Students from Peru and Paraguay exhibited the highest and most homogeneous levels of entrepreneurial attitude. This suggests that the educational or cultural context may be actively fostering favorable dispositions toward entrepreneurship. In contrast, Mexico and Ecuador showed greater dispersion in responses and a higher proportion of students with low or moderate levels of entrepreneurial attitude, particularly among those in the early semesters. Colombia, on the other hand, displayed high levels of entrepreneurial attitude, though with lower density, indicating a more heterogeneous presence of this trait within the sample.

University support

Perception of university support was examined through the lens of training. Figure 4 illustrates how students’ perceptions of university-based entrepreneurship education vary according to their prior entrepreneurial experience. Overall, in countries such as Peru and Colombia, both students with and without entrepreneurial experience rated the training received positively, with high medians and low dispersion, suggesting the presence of well-established educational programs.

Figure 4. Perceived university support by entrepreneurial experience.

Raincloud plots comparing perceived entrepreneurship education between students with and without prior entrepreneurial experience. Grouped by country; scores on a 5-point Likert scale.

In contrast, in Mexico, Ecuador, and Paraguay, although median scores remain within moderate ranges, the dispersion is greater, and no substantial differences are observed between students with and without entrepreneurial experience. This may indicate that, while entrepreneurship education is present, it does not consistently translate into a differentiated perception of its usefulness or applicability.

Self-efficacy

On the other hand, we examined whether actual entrepreneurial experience reinforced perceived self-efficacy within the study sample ( Figure 5). The figure suggests that entrepreneurial expertise is associated with higher levels of perceived self-efficacy across all countries represented.

Figure 5. Entrepreneurial self-efficacy by entrepreneurial experience.

Raincloud plots comparing self-efficacy between students with and without prior entrepreneurial experience, grouped by country. Scores on a 5-point Likert scale.

Peru and Mexico show a more apparent distinction between students with and without entrepreneurial experience: those with experience exhibit higher levels of self-efficacy. Ecuador, Paraguay, and Colombia also reflect this trend, although with a smaller gap between groups, suggesting that expertise still contributes, but may be interacting with other contextual factors.

Cognitive load

On the other hand, we found that students in advanced semesters tend to experience higher cognitive load ( Figure 6), particularly in Peru, Mexico, and Paraguay. This is to be expected, as academic demands increase, extracurricular commitments and vocational decisions become more frequent, and personal and societal expectations regarding professional futures intensify.

Figure 6. Cognitive load by academic semester and country.

Raincloud plots showing cognitive load scores by semester for each country. Higher scores indicate greater perceived load. 5-point Likert scale.

Peru and Paraguay show greater dispersion across semesters, allowing for a more precise visualization of the progression of cognitive stress. In Ecuador and Colombia, the trend is less pronounced, possibly due to differences in curricular load or pedagogical approach.

Emotional well-being

Regarding students’ perceived emotional well-being, Figure 7 shows that, in most countries, students with prior entrepreneurial experience tend to report slightly higher or more consistent levels of well-being compared to those without experience. In Peru, students with entrepreneurial experience are more concentrated at higher levels of well-being, with lower dispersion than their inexperienced peers, which may be associated with greater self-fulfillment or a stronger sense of entrepreneurial purpose.

Figure 7. Emotional well-being by entrepreneurial experience.

Raincloud plots comparing well-being between students with and without prior entrepreneurial experience. Scores on a 5-point Likert scale; higher scores indicate higher well-being.

In Mexico and Ecuador, although there is some overlap, a slight advantage in well-being can also be observed among experienced students, although the differences are not pronounced. In Paraguay, the high dispersion in both groups is notable, especially among those without experience, which may reflect greater uncertainty or lack of direction. Finally, in Colombia, both groups show similar levels, although students with experience appear less likely to report very low levels of emotional well-being.

Flow

Figure 8 shows an upward trend in work-related flow across academic semesters, which can be interpreted as a progressive increase in students’ ability to organize tasks, maintain focus, and master content as they advance in their university education. This relationship appears to be more evident in Peru and Mexico, where students in higher semesters report higher levels of flow compared to those in earlier semesters.

Figure 8. Flow levels by academic semester and country.

Raincloud plots showing flow scores across semesters in each country. Scores on a 5-point Likert scale.

In countries such as Ecuador, Paraguay, and Colombia, the relationship between semester and flow level appears less clear, with a wider dispersion of responses across different academic levels. This may be due to differences in pedagogical styles, institutional support, or curricular load, which warrants further examination.

Emotional competence

Emotional competence was examined across three dimensions: self-awareness, empathy, and social skills. In the case of self-awareness ( Figure 9), several countries—notably Mexico and Colombia—show a greater concentration of high scores in mid and advanced semesters, which may indicate that prolonged exposure to university experiences fosters the development of self-awareness.

Figure 9. Self-awareness by academic semester and country.

Raincloud plots showing self-awareness scores across semesters for each country. Scores on a 5-point Likert scale.

In Peru and Ecuador, scores are more dispersed, but a slight upward trend is observed in advanced semesters. In Paraguay, while there is greater overall dispersion, self-awareness levels appear to remain relatively high even in the early semesters, which may be related to specific characteristics of the academic program or type of training provided.

Regarding empathy ( Figure 10), the data suggest that entrepreneurial experience is associated with slightly higher levels of empathy in several national contexts. This is particularly evident in Mexico, Paraguay, and Colombia, where participants with experience show score concentrations closer to the maximum. In Peru, empathy levels are generally high across all groups, with no notable difference between those with or without experience. In Ecuador, both groups show greater dispersion, although a slight advantage is observed among students with entrepreneurial experience.

Figure 10. Empathy by entrepreneurial experience and country.

Raincloud plots comparing empathy levels between students with and without entrepreneurial experience. Scores on a 5-point Likert scale.

This may be explained by the fact that entrepreneurship exposes students to situations where they must understand and anticipate the needs of others, develop effective communication, and collaborate—skills closely related to empathy.

Regarding social skills ( Figure 11), although no drastic trend is observed, in some countries such as Peru and Colombia, there appears to be a slight positive association between academic semester and social skills. In other words, as students progress through their university education, they may be developing social skills more frequently—such as teamwork, leadership, and interpersonal relationship management—likely due to increased exposure to group experiences, presentations, internships, or collaborative projects.

Figure 11. Social skills by academic semester and country.

Raincloud plots showing social skills scores across semesters in each country. Scores on a 5-point Likert scale.

However, this trend is not consistent across all countries, which may be due to curricular or institutional differences in the type of training and activities that foster these skills.

Entrepreneurial action

Entrepreneurial action was assessed through two dimensions: discovery and exploitation. Regarding the discovery phase, Figure 12 suggests that students with prior entrepreneurial experience tend to exhibit higher levels of entrepreneurial action compared to those who have not previously engaged in entrepreneurship.

Figure 12. Entrepreneurial action – discovery phase by entrepreneurial experience.

Raincloud plots showing discovery-phase entrepreneurial action scores for students with and without prior entrepreneurial experience, by country. Scores on a 5-point Likert scale.

In countries such as Mexico, Ecuador, and Colombia, a greater concentration of high scores is observed among students with entrepreneurial experience. In Peru, although overall entrepreneurial action is high, the distribution is also elevated for the experienced group. In Paraguay, the difference is less pronounced, but the general trend remains consistent.

A similar finding emerges in the exploitation dimension ( Figure 13), where students with entrepreneurial experience tend to report slightly higher levels of entrepreneurial action—particularly in Mexico and Ecuador. However, in countries like Paraguay and Colombia, the differences are much less pronounced, and both distributions (with and without experience) show greater dispersion and overlap.

Figure 13. Entrepreneurial action – exploitation phase by entrepreneurial experience.

Raincloud plots showing exploitation-phase entrepreneurial action scores for students with and without prior entrepreneurial experience, by country. Scores on a 5-point Likert scale.

In the case of Peru, although students with entrepreneurial experience show higher scores, there is also a group without experience that exhibits similarly high levels. This may indicate that the educational environment or local entrepreneurial ecosystem supports the transition to the exploitation phase even without prior experience.

Financial knowledge

Finally, regarding financial knowledge ( Figure 14), Peru presents the highest and most consistent levels among the countries analyzed, with a median close to 4.5 and a distribution concentrated in the upper range of the scale. Both students with entrepreneurial experience and those without it tend to perform well on this variable, suggesting a potentially more effective integration of financial content into the Peruvian university curriculum.

Figure 14. Financial knowledge by entrepreneurial experience and country.

Raincloud plots comparing financial knowledge scores between students with and without entrepreneurial experience. Higher scores indicate better financial knowledge.

In contrast, Mexico shows greater dispersion in responses, including cases with very low levels of financial knowledge, which may reflect disparities in training or access to basic economic content. Ecuador, Paraguay, and Colombia show moderate medians (between 3.5 and 4), but also higher variability—especially in Paraguay—possibly indicating gaps in formal financial education or limited practical exposure to fundamental economic concepts.

Results of differences between countries

Since the variables did not meet the assumptions of normality or homogeneity of variances, the non-parametric Kruskal-Wallis test was used to determine whether significant differences existed among the five participating countries (Peru, Mexico, Ecuador, Paraguay, and Colombia) in key variables related to university entrepreneurship. Of the 18 variables analyzed, nine showed statistically significant differences (p < 0.05), as summarized in Table 1.

To verify empirical adequacy, a post hoc power analysis was conducted for an omnibus comparison equivalent to the Kruskal–Wallis test (five groups, α = .05, medium effect f = 0.25 ≈ η² = 0.06; Cohen, 1988). The resulting statistical power was 0.97, indicating that the total sample was more than sufficient to detect effects of medium magnitude. Consistently, the descriptive results (Alba et al., 2025) show low standard errors (SE ≤ 0.14) and small to moderate effect sizes (ω² and ε²), supporting the precision and stability of the estimates.

The results clearly show that Mexico ranks highest in several key dimensions, particularly in university training, financial knowledge, entrepreneurial action, and emotional well-being. Peru also stands out for exhibiting the highest levels of entrepreneurial intention compared to the other countries. In contrast, Ecuador and Paraguay tend to display lower scores across multiple dimensions, suggesting potential gaps in the development of the university entrepreneurial ecosystem.

Correlations with entrepreneurial intention and action

To assess the factors associated with entrepreneurial behavior, Spearman’s rank correlation (ρ) was applied, allowing for the identification of the strength and direction of associations between ordinal variables without assuming normality. The results reveal consistent associations between entrepreneurial intention (PromEI) and key variables of the entrepreneurial profile. Entrepreneurial attitude (PromEA) shows the strongest correlation with PromEI (ρ = 0.664, p < .001), followed by self-efficacy (PromSE, ρ = 0.370), entrepreneurial action in the discovery phase (PromEActD, ρ = 0.371), and perceived university support (PromU, ρ = 0.288). These findings suggest that strengthening personal beliefs, attitudes, and institutional support is linked to higher levels of entrepreneurial intention.

Regarding entrepreneurial action in the discovery phase (PromEActD), it is strongly associated with the exploitation phase (PromEActE, ρ = 0.642), self-efficacy (PromSE, ρ = 0.507), financial knowledge (PromFK, ρ = 0.397), and self-awareness (PromSA, ρ = 0.353). These relationships suggest that students who are already acting on opportunities also tend to discover new ones, particularly when they feel capable and possess the cognitive and emotional tools to evaluate their environment.

Finally, in the exploitation phase (PromEActE), in addition to its strong association with the discovery phase (ρ = 0.642), significant correlations were observed with flow (PromFL, ρ = 0.379), self-efficacy (PromSE, ρ = 0.283), emotional well-being (PromEW, ρ = 0.275), and financial knowledge (PromFK, ρ = 0.252). These results reinforce the idea that successfully executing entrepreneurial projects requires not only technical preparation but also emotional stability and personal confidence.

Regression models for entrepreneurial intention and action

The multiple linear regression models were statistically significant and explained a substantial proportion of variance in their respective dependent variables. The model with PromEI (entrepreneurial intention) as the dependent variable showed a good fit (F(5, 534) = 133.72, p < .001), explaining 55.2% of the variance (adjusted R² = 0.552). The model for PromEActD (entrepreneurial action – discovery) showed the highest fit (F(3, 536) = 269.22, p < .001), accounting for 59.9% of the variance (adjusted R² = 0.599). Meanwhile, the model for PromEActE (entrepreneurial action – exploitation) was also significant (F(5, 534) = 26.14, p < .001), explaining 27.3% of the variance (adjusted R² = 0.273). In all cases, the classical assumptions of linear regression were verified, and no multicollinearity issues were identified.

In the PromEI model ( Figure 15), entrepreneurial attitude (PromEA) emerged as the strongest predictor (β ≈ 0.71), highlighting its key role as a transversal formative axis. It was followed by entrepreneurial action – discovery (PromEActD) (β ≈ 0.14) and self-efficacy (PromSE) (β ≈ 0.12), underscoring the importance of teaching students how to identify real opportunities and build entrepreneurial confidence. University support (PromU) also had a small but significant effect (β ≈ 0.08), indicating that the institutional environment should be reinforced. Finally, empathy (PromE) showed an adverse impact (β ≈ –0.10), possibly suggesting that in specific contexts, high levels of empathy may be associated with lower entrepreneurial intention—perhaps due to risk aversion or a less autonomous orientation.

Figure 15. Multiple linear regression model predicting entrepreneurial intention.

Path diagram showing standardized beta coefficients for predictors of entrepreneurial intention (PromEI).

For PromEActD ( Figure 16), the three significant predictors were: entrepreneurial action – exploitation (PromEActE) (β ≈ 0.56), which emerged as the strongest factor, suggesting that teaching students how to execute projects enhances their ability to discover opportunities. Self-efficacy (PromSE) (β ≈ 0.26) and self-awareness (PromSA) (β ≈ 0.23) also showed significant effects, highlighting the importance of cultivating both personal confidence and the ability to recognize and regulate one’s own emotions and strengths during the early stages of the entrepreneurial process.

Figure 16. Regression model for entrepreneurial action – discovery phase.

Path diagram showing significant predictors of discovery-phase entrepreneurial action (PromEActD).

Regarding PromEActE ( Figure 17), the model revealed five significant predictors. Flow (PromFL) stood out with a relevant positive effect (β ≈ 0.22), followed by self-efficacy (PromSE) (β ≈ 0.19), practical financial knowledge (PromFK) (β ≈ 0.15), and emotional well-being (PromEW) (β ≈ 0.11), pointing to the importance of preparing students who are not only technically skilled but also emotionally balanced. However, cognitive load (PromCL) showed an adverse effect (β ≈ = -0.25), which may indicate that excessive mental demands or academic overload interfere with students’ ability to carry out entrepreneurial actions.

Figure 17. Regression model for entrepreneurial action – exploitation phase.

Path diagram showing significant predictors of exploitation-phase entrepreneurial action (PromEActE).

Recommendations

Based on the study’s findings, the following recommendations are proposed to strengthen the university’s entrepreneurial support ecosystem. These recommendations integrate strategic institutional aspects as well as curricular and operational components of academic programs, acknowledging the influence of contextual, emotional, and cognitive variables that vary across countries.

Strategic Recommendations (Institutional and Curricular Level)

The cross-country differences in entrepreneurial intention, attitude, and motivation highlight the need for adaptive curricula that are responsive to students’ cultural and emotional profiles. In Ecuador, it is recommended to include activities that promote proactivity and entrepreneurial vision, such as social innovation projects or entrepreneurship challenges. In Mexico, emotional well-being and self-regulation could be incorporated transversally into the entrepreneurship curriculum through reflective practices or stress management workshops.

Given that empathy, self-efficacy, and motivation were highly correlated with entrepreneurial intention, it is advisable to integrate these competencies as explicit learning outcomes in key courses. This can be achieved through active methodologies such as project-based learning, role-plays, ethical decision-making simulations, cooperative activities, and peer feedback and self-assessment sessions.

The impact of social support on both entrepreneurial intention and action justifies the design of collaborative training spaces such as entrepreneurship labs or academic incubators, where interaction, teamwork, tolerance for failure, and communities of practice are fostered. These spaces could be aligned with academic credit or incorporated into practical courses.

The finding that cognitive load negatively affects the entrepreneurial exploitation phase (β = –0.25) suggests the need to revise the sequencing and density of entrepreneurship courses, avoiding their concentration during high-demand semesters. It is recommended to incorporate into the curriculum short sessions on mindfulness, study techniques, and emotional regulation—or even active break weeks—to help prevent cognitive and emotional exhaustion.

Moreover, regression results showed that entrepreneurial attitude (β = 0.71) and self-efficacy (β ≈ 0.12–0.26) are the most relevant predictors of entrepreneurial intention and action. Therefore, it is essential to reinforce these variables across the curriculum through leadership workshops, entrepreneurial storytelling, practical challenges, and spaces for individualized feedback. In addition, self-awareness—which showed a positive effect in early stages—should be promoted through reflective portfolios and metacognitive exercises.

Operational Recommendations (Entrepreneurship Programs and Student Support)

It is recommended to offer personalized training pathways according to the student’s stage in the entrepreneurial process:

The correlations found between self-regulation, motivation, and self-awareness support the development of diagnostic tools to be applied at the beginning of the academic cycle (e.g., surveys on emotional competencies and entrepreneurial profiles). These results can be used to assign tailored mentoring, suggest electives, and personalize students’ learning pathways. Additionally, it is recommended to monitor indicators of cognitive load and emotional well-being during the later stages of academic training.

It is also advisable to include mentorship spaces with entrepreneurial alums or emotional coaching sessions led by professionals, focused on strengthening self-efficacy, resilience, and sense of purpose. These activities can be formally integrated into capstone projects or supervised internships, with the possibility of earning academic credit.

Finally, implement continuous evaluation systems that assess not only academic outcomes but also the evolution of emotional well-being, perceived institutional support, and the progression of the entrepreneurial profile. This information will support curriculum refinement, adjustment of academic workload, and better targeting of institutional resources.

Practical implications derived from the evidence

Because attitude is the dominant predictor of intention and self-efficacy, flow, and financial knowledge support action—while cognitive load constrains exploitation—universities seeking to strengthen entrepreneurial outcomes should: prioritize experiential, mentored learning that cultivates positive attitudes (Lackéus, 2020).

Design efficacy-building sequences with progressive mastery and feedback (Bandura, 1991); embed practical financial content and project-execution training to facilitate exploitation (Shane & Venkataraman, 2000); and actively manage academic load and teach self-regulation to protect students’ cognitive resources during project implementation.

Limitations and avenues for future research

Although the results are robust across multiple techniques, the gender–risk pattern was not statistically significant and should be tested with larger, possibly longitudinal samples (Croson & Gneezy, 2009). More broadly, future research should adopt longitudinal designs to observe how attitudes, efficacy, and competencies evolve over time and how cognitive load fluctuates around key educational milestones (Kautonen et al., 2015). Cross-national designs should also examine how institutional supports moderate the TPB pathways and the competency–action linkage posited here.