This study adopted two vital theories such as the theory of planned behaviour (TPB) (Ajzen 1991) and the social learning theory (SLT) (Bandura 1977) to explain the association among variables in the conceptual model. The TPB was developed by Ajzen (1991) to explain the development of EI and actual behaviour. Ajzen (1991) posits that EI results from the interaction of three crucial variables such as attitude, subjective norms and perceived behavioural control. Attitude describes an individual’s perception of whether it is favourable to engage in a certain behaviour (Ajzen 1991). This means that the expectations and beliefs about the benefits and challenges of an entrepreneurial career are the basis for an individual’s attitude towards engaging in related behaviour. Subjective norms describe whether the individual will act because of the influence of those around them. Thus, individuals are likely to engage in certain behaviour if they believe that those around them support them to do so (Ajzen 1991). Subjective norms are also often associated with societal pressure on the individual to engage in a certain behaviour deemed socially desirable (Ajzen 1991). Perceived behavioural control is referred to as an individual’s judgement of their capacity to carry out a specific behaviour (Ajzen 1991). The level of control a person has over their behaviour determines their capability to act and predict future behaviour (Ajzen 1991). The TPB also postulates that EI precedes EB. This suggests that strong intentions propel individuals to engage in EB (Ajzen 1991). The TPB has been adopted widely in existing studies in entrepreneurship, demonstrating its usefulness as a theoretical framework to explain the formation of EI and actual behaviour.

We further adopted the SLT (Bandura 1977) to theorise the inclusion of role models in the conceptual model of this study. According to Bandura (1977), individuals can learn not only directly but also by observing the actions of others and the consequences that follow in social settings. Bandura (1977) highlighted four key processes involved in observational learning: attention (noticing the behaviour), retention (remembering it), reproduction (being able to replicate it) and motivation (having a reason to imitate it). According to Bandura (1977), an individual’s attention is drawn from observing successful individuals around them who become their role models. The retention processes encompass the recollection of an experience or behaviour over a specific period. To ensure a skilful mastery of observational learning, it is crucial to evaluate and reproduce the observed behaviour (Bandura 1977). The process of reproduction is a complex one, requiring both mental and physical faculties to mimic observed behaviour while performing a given task (Bandura 1977; Johnson 2018). Lastly, Bandura (1977) argues that motivation is primarily driven by human agency rather than external factors such as positive rewards. Once individuals have identified their role models, they are motivated to engage in certain behaviour. Furthermore, Bandura indicated that the issues of reinforcement and punishment also model the behaviour that can be learned from role models. Vicarious reinforcement involves a person’s behaviour changing as a result of observing another person’s behaviour being reinforced, without being actively reinforced themselves (Bandura 1977). This process is a behavioural mechanism that causes an observer to behave more like a model because of the model’s behaviour being reinforced or punished (Bandura 1977). The SLT is a vital theory for this study because it helps to explain the role of environmental factors (role models) in the conceptual model of this study. The SLT has also been widely used in entrepreneurship studies across different contexts.

Integrating the TPB and SLT provides a novel and comprehensive framework for understanding the determinants of EB among students. While the TPB explains how EI predicts behaviour, existing studies relying solely on the TPB to predict behaviour have been criticised for overlooking the contextual factors that influence individuals’ translation of their intentions into action (Duong 2023; Tsou et al. 2023). By incorporating SLT, which emphasises learning through observation and imitation of role models, this study introduces a crucial social dimension into the TPB framework from a developing country context. This integration acknowledges that, beyond internal cognitive drivers, students’ engagement in entrepreneurial activities is shaped by observing others who have successfully engaged in entrepreneurship, thereby enhancing self-efficacy and reducing fear of failure. In developing contexts such as South Africa, where entrepreneurial engagement among students remains low because of several challenges, including weak entrepreneurial framework conditions (GEM 2025), role models can inspire and motivate students to persist in implementing their entrepreneurial goals, such as starting a business. Therefore, combining TPB and SLT offers a theoretical perspective that helps bridge gaps in the existing literature by identifying role models as a boundary condition that can strengthen or weaken the EI–behaviour relationship from a developing country context. This approach enhances our understanding of why some students act on their EIs while others do not (Abbasianchavari & Moritz 2021), thereby providing a more contextually grounded explanation relevant to developing economies, such as South Africa.

The association between EI and actual behaviour is crucial in the entrepreneurial domain (Frese & Gielnik 2023; Waheed, Laeequddin & Sahay 2025). This is because it forms the foundation of how an individual starts with an entrepreneurial goal, which is then implemented to start a business (Frese & Gielnik 2023). According to Ajzen (1991), EI is a precursor to EB. This suggests that for one to engage in entrepreneurship, they could have thought about it for some time (Cui & Bell 2022). Other studies also support the view that EI is a crucial determinant of entrepreneurial activities (Duong 2022; Neneh & Dzomonda 2024). Therefore, positive EI is required to enhance actual behaviour (Ajzen 1991). Mackiewicz (2023) argues that new venture creation is an intentional activity rather than an unplanned event. Engaging in EB demonstrates the ability to pursue the initial goal and engage in actual activities crucial for new venture creation (Duong 2022). According to Kong et al. (2020), individuals with high EIs are more likely to establish a new business than those with weak EI. This is because individuals with high EIs are more willing to take risks, embrace uncertainty and persist in the face of challenges. Existing literature explains that EI is not merely a desire or wish to start a business, but a cognitive state that precedes EB (Naskar & Lindahl 2025; Waheed et al. 2025). It represents an individual’s commitment to starting a business and reflects their readiness to act towards realising their entrepreneurial goals (Gieure, Benavides-Espinosa & Roig-Dobón 2020; Saoula et al. 2025; Tran et al. 2024). Extant scholarly work suggests that EI is a crucial determinant for actual behaviour (Duong 2022; Frese & Gielnik 2023; Neneh 2019; Neneh & Dzomonda 2024). These studies confirm the foundations of the TPB, which suggests that without an intention, it is difficult to track the development of a particular behaviour (Naskar & Lindahl 2025). Existing studies (Tsou et al. 2023; Zhuang & Sun 2025) argue that although the EI–behaviour relationship is often established, inconsistencies exist in the findings. The findings on the effect of EI on behaviour range from negligible to large. While a study by Zhuang and Sun (2025) found that in some instances, EI is reported to account for between 20% and 70% of variance in actual behaviour, a study by Tsou et al. (2023) reported that EI accounted for only 17% of variance in actual behaviour. These inconsistencies necessitate more empirical studies to bridge these gaps. Understanding the association between EI and actual behaviour is particularly important in the context of South Africa, where the government is actively seeking alternative ways to create employment to address the country’s various socio-economic challenges (GEM 2025). On this account, this study hypothesises that:

Role models play a crucial role in the entrepreneurial process (Nucci et al. 2025). Thus, role models can help an individual translate their EI into behaviour (Kong et al. 2020). Entrepreneurial role models offer personal motivations and behavioural guidance, which has a significant influence on EB (Nucci et al. 2025). Existing literature suggests that there are different types of role models that influence individuals differently in their decision to become entrepreneurs (Abbasianchavari & Moritz 2021; Kong et al. 2020; Nucci et al. 2025). Based on this account, it becomes crucial to understand the role of each type as a necessary boundary condition to understand the EI–EB relationship.

Existing studies assert that successful entrepreneurial parents can inspire an individual to consider an entrepreneurial career through socialisation (Nucci et al. 2025; Rajchamaha & Prapojanasomboon 2022). Parents and immediate family are the first sphere closely linked to childhood development. As such, what an individual observes as they grow becomes part of their reality and shapes career choices, such as becoming an entrepreneur (Nucci et al. 2025). In support, Maleki et al. (2023) assert that the immediate family has an impact on an individual’s behaviour, values, personal growth and development. This suggests that individuals born into an entrepreneurial family are more likely to develop an intention and subsequently behave in a manner consistent with starting a business (Roos & Botha 2022; Xuan & Yankai 2025). Successful entrepreneurial parents can serve as a source of inspiration for their children to consider venturing into entrepreneurship (Haque, Kour & Mahin 2024). Engagement in EB is influenced by family, as having an entrepreneurial family can foster an entrepreneurial identity among children, enabling them to learn the dynamics of running a business at a younger age (Bagherian, Soleimanof & Feyzbakhsh 2025; Roos & Botha 2022). This implies that individuals with a high exposure to successful entrepreneurial family members will likely engage in entrepreneurial activities than those with low exposure. Whether this effect is uniform across developed and developing country contexts remains indistinct. On this account, the hypothesis is stated as:

Knowing successful entrepreneurs enhances one’s confidence in becoming an entrepreneur (Boldureanu et al. 2020; Bosma et al. 2012). According to Boldureanu et al. (2020), a successful entrepreneurial role model can assume various roles that help the potential entrepreneur navigate all stages of the business life cycle. In this case, a successful entrepreneurial role model can assume the role of a coach throughout the stages of the entrepreneurial venture, inspiring and intellectually stimulating the nascent entrepreneur to think creatively, thus helping them navigate the challenging entrepreneurial career (Boldureanu et al. 2020). Ideally, having someone successful in business can increase one’s propensity and confidence to start their own business (Nucci et al. 2025). Successful entrepreneurs inspire and motivate individuals to engage in entrepreneurial activities (Abbasianchavari & Moritz 2021). According to Kong et al. (2020), individuals who can easily interact with successful role models tend to exhibit lower levels of fear of failure, which triggers them to engage in entrepreneurial activities. This is particularly important in the context of South Africa, characterised by a challenging entrepreneurial environment with weak institutional support for young entrepreneurs (GEM 2025). Thus, exposure to entrepreneurs who have built successful businesses in the same environment and context can help shape the entrepreneurial mindset of students, helping them transition from EI to actual behaviour.

Furthermore, existing literature reports that individuals with high exposure to SERM tend to engage in actual behaviour more than those with low exposure (Abbasianchavari & Moritz 2021; Kong et al. 2020; Nucci et al. 2025). Although existing literature enhances our understanding of the impact of SERM on the entrepreneurial process, there are limited studies that have explored the moderating role of SERM in the EI–behaviour relationship in the context of South Africa. On this background, this study hypothesises that:

Scholars and policymakers have shown a keen interest in understanding the influence of peers on individuals’ decisions to become entrepreneurs (Portyanko et al. 2023). Existing studies show that peers influence one another to pursue certain careers, such as entrepreneurship (Lingappa, Shah & Mathew 2020; Portyanko et al. 2023). This phenomenon is mostly common among university peers who are spatially close to each other. Thus, peers may have a greater influence on others regarding entrepreneurial endeavours and achievements than formal institutional structures or policies (Abbasianchavari & Moritz 2021). Specifically, such role models increase the likelihood of students choosing a business-related major and deciding to pursue business interests after graduation (Bosma et al. 2012; Portyanko et al. 2023). This suggests that students with exposure to peers who successfully run a business are more likely to opt for entrepreneurial activities to start their own business than those without successful PR. To further test this, the following hypothesis is proposed:

Educators play a crucial role in influencing the EB of students (Abbasianchavari & Moritz 2021). According to Machingambi and Iwu (2025:4), ‘ideal entrepreneurship educators need to guide learners through the four-stage process, that is, concrete experience, reflective observation, abstract conceptualisation, and active experimentation’. Prastyaningtyas et al. (2023) emphasise the pivotal role played by educators in developing and enhancing students’ entrepreneurial skills. The study found that through their guidance and instruction, educators can help students develop the knowledge, mindset and confidence needed to identify and pursue opportunities. A study by Iwu et al. (2021), conducted with a sample of university students in South Africa, found that the competencies of entrepreneurship education educators have a positive influence on EI. Nevertheless, another study by Diegoli et al. (2018) found that the educator’s business experience was not a significant determinant of engagement in entrepreneurial activities. Be that as it may, existing literature reports that students with access to educators who have real-world experience in running a business tend to find it easier to translate their EI into actual behaviour (Abbasianchavari & Moritz 2021; Machingambi & Iwu 2025; Zhou, Rashid & Cheng 2024). This is because educators with previous business experience find it easier to integrate experiential learning into their teaching methods, thereby boosting the students’ confidence in their entrepreneurial capabilities. This is particularly important in the context of South Africa, where students display high fear of failure and low entrepreneurial drive to start a business. Thus, these educators can help students translate their EI into behaviour by nurturing their entrepreneurial mindset, which values experimentation, adaptability and opportunity recognition and exploitation in a challenging entrepreneurial environment, such as South Africa. On this backdrop, the study proposes to test the following hypothesis:

The respondents were asked to indicate their level of disagreement or agreement on a statement anchored on a five-point Likert scale ranging from (1) strongly disagree (SD) to (5) strongly agree (SA).

To account for potential alternative explanations for actual behaviour, the study also included control variables such as age, gender, family business background (FBB) and prior business experience. Existing studies indicate that these factors influence EB (Shirokova et al. 2016).

Data were analysed using SPSS and Smart-PLS 4. Specifically, the SPSS software was used to compute descriptive statistics such as mean and standard deviations as well as correlations among variables. Partial least squares structural equation modelling (PLS-SEM) using SMART-PLS version 4 was used to test the hypothesised relationships regarding the influence of entrepreneurial role models on the link between EIs and EB. Partial least squares structural equation modelling was particularly well-suited for this study as it enabled the researchers to examine a complex model that incorporated the moderating effects of multiple entrepreneurial role models on the EI–EB relationship.

Ethical clearance to conduct this study was obtained from the University of the Free State, General/Human Research Ethics Committee (No. UFS-HSD2023/0892). This study adhered to all ethical guidelines in research involving humans as participants. The study obtained ethical clearance from the university ethics committee before commencing with data collection. All the participants in the study provided informed consent to voluntarily participate in the study.

In this study, the findings showed that the majority of the participants (65%) were females while 35% were males. These participants were mainly aged 18–21 years (67.6%) followed by those aged 22–25 years (29.9%), clearly demonstrating the general age distribution of undergraduate students in South African universities. The findings further showed that the majority of the surveyed students were black people (91%), followed by those who identified themselves as Coloured people (6%) and lastly white people (3%). In terms of prior business experience, 75% of the participants indicated that they did not have any prior business experience, while only 25% of the participants agreed that they had previously operated a business.

Table 1 presents results on the descriptive statistics and correlations among the key variables of the study. The findings show that the students possess a moderate level of EI (mean = 3.74; SD = 1.14), with low EB, as indicated by an average mean of 2.70. This implies that the university students surveyed are not actively engaging in entrepreneurial activities. In terms of exposure to different role models, the findings showed that participants had moderate exposure to PRM (mean = 3.75) and PR (mean = 3.63). On the other hand, the results showed that participants had a weak exposure to successful entrepreneurs as role models (mean = 3.37) and educator role models (mean = 2.91). In terms of correlations, the findings showed that EI was positively correlated with EB (r = 0.496; p < 0.01). The findings further showed that EI was positively correlated with PRM (r = 0.296; p < 0.01), SERM (r = 0.508; p < 0.01) and PR (r = 0.342; p < 0.01). Nevertheless, the correlation between EI and educators as role models (EM) was insignificant (r = 0.029; p > 0.05). Lastly, the findings showed that EB was positively correlated with PRM (r = 0.249; p < 0.01), SERM (r = 0.502; p < 0.01), PR (r = 0.332; p < 0.01) and EM (r = 0.254; p < 0.01).

According to Hair et al. (2019), the first step when using SMART-PLS is to evaluate the quality of the measurement model. This is intended to assess the extent to which adapted constructs accurately and consistently measure the intended concepts (Hair et al. 2019). The results show that all the factors loaded accurately. As shown in Table 2 above, the factor loadings range from 0.712 to 0.922, which is acceptable. The assessment of reliability for the scales was performed using Cronbach’s alpha (CA) and composite reliability. The CA values ranged from 0.703 to 0.914, while composite reliability values ranged from 0.740 to 0.933 indicating that the scales were reliable. These values were above the recommended threshold of 0.70 as suggested by Hair et al. (2019). To evaluate the convergent validity of the constructs, the researchers used the AVE. The study’s AVE values ranged from 0.553 to 0.746, indicating satisfactory levels of convergent validity.

Table 3 presents discriminant validity results. We adopted the heterotrait–monotrait (HTMT) ratio of correlations to assess discriminant validity. The HTMT is considered a more precise measure for discriminant validity compared to other existing criteria, such as the Fornell–Larcker criteria (Hair et al. 2019; Henseler, Hubona & Ray 2016). Using this criterion, discriminant validity is achieved when the values are below 0.9 or within a more conservative value of 0.85 (Hair et al. 2019). In this study, discriminant validity was achieved as all the HTMT values fell within the recommended threshold.

According to Hair et al. (2019), the structural model should be assessed for its explanatory power by evaluating the R² values. Additionally, the structural model should be evaluated to ensure that it is free from collinearity issues by assessing the variance inflation factor (VIF) values of the constructs (Hair et al. 2019). In this study, the model R² was 0.38. The R² value of 38% represents the proportion of the variation in the dependent variable that is explained by the independent variables in the model. In other words, the model accounts for 38% of the actual behaviour. In this study, the VIFs for EI, EM, SERM, PRM, PR and EB were 1.726, 2.813, 1.664, 1.445, 2.276 and 2.239, respectively. These values are within the acceptable threshold of 3 and below, indicating that there were no collinearity issues in the model (Hair et al. 2019). In this study, the predictive relevance of the model was assessed using the Q². The Q² for the endogenous variable was 0.522, which was above zero, suggesting an acceptable predictive relevance of the model. The significance of the hypothesised relationships was tested using the bootstrapping method with 5,000 subsamples in SMART-PLS 4. As shown in Table 4, the results show that EI significantly predicts EB (β = 0.476; t = 10.317; p = 0.000).

Table 5 presents findings on the moderating role of different types of entrepreneurial role models, including PRM, SERM and PR, on the association between EI and EB.

As shown in Table 5, the results showed that only SERM positively and significantly moderate the association between EI and actual behaviour (β = 0.090; t = 2.259; p = 0.023). This supported hypothesis 2b. The findings imply that SERM play a significant role in positively influencing university students to translate their EI into actual EB. On the other hand, the results showed that other types of role models, such as educators, parental and PR, are insignificant moderators of the EI–EB relationship in this study. Therefore, hypotheses H2^a, H2^c and H2^d were rejected.

Even though this study achieved its objectives, there are some limitations worth noting. The study was conducted using a sample of university students from one faculty, which may limit the generalisability of the findings. In addition, the convenience sampling technique adopted in this study may limit the generalisability of the findings. Furthermore, the study was cross-sectional, limiting a nuanced understanding of how individuals translate their EI into EB over time. The study was also limited only to the impact of role models on the EI–EB link. It is essential to note that various other factors may influence this relationship. Thus, further studies may consider conducting a longitudinal study with similar variables using a sample that covers all the universities in South Africa to enhance the generalisability of the findings. As such, future studies may consider expanding the scope of this study by investigating other contextual factors present in the entrepreneurial ecosystem, such as culture, support and the availability of markets, as potential boundary conditions necessary to understand the EI–EB relationship.