The current study adopts an explanatory approach, utilizing quantitative research through closed-ended questionnaire surveys with 5-point Likert scale responses, with 1 for strongly disagree to 5 for strongly agree. The online questionnaires were distributed across seven universities, comprising four private and three public institutions in Pakistan. The proposed mediating model consists of five constructs: entrepreneurial inspiration, entrepreneurial skills, entrepreneurial awareness, entrepreneurial education, and entrepreneurial intention. The constructs of the model were evaluated through different items from different resources, such as entrepreneurial inspiration, entrepreneurial skills, and entrepreneurial awareness, which have four items. Entrepreneurial education and entrepreneurial intention have five dedicated items to construct for evaluation. Data were gathered from final-year business students in the 2020 and 2021 cohorts at seven universities in the Punjab region of Pakistan (see
Table 2). To analyze the path relationships among the constructs and test the hypotheses, SmartPLS (version 4) was employed.
The research population consists of 867 final-year business students from the 2020 to 2021 batches, specifically targeting those who had completed an entrepreneurship course. A simple random sampling technique was employed to gather data from final-year business students from various Pakistani universities in the 2020 to 2021 batches, specifically those who are willing to participate in the research. A total of 900 questionnaires were distributed, and a total of 886 responses were received from the respondents. After screening the data, 865 responses were deemed suitable for further analysis, with 18 respondents excluded due to blank responses and straight-line responses.
3.2. Model Measurement
According to
Al-Qadasi et al. (
2023), various factors contribute to students’ entrepreneurial intentions and foster their entrepreneurial efficacy. These variables include personal input, environmental input, and behavioral output, which motivate individuals to pursue an entrepreneurial path and determine their future objectives. Further,
Shahzad et al. (
2021) proposed that the availability of entrepreneurship education considerably impacts students’ entrepreneurial intentions. Therefore, to measure the construct of the model, such as to evaluate entrepreneurial inspiration, four items were used and adopted from
Li et al. (
2023), four items were adopted from
Kisubi et al. (
2021) for the evaluation of entrepreneurial skills, four items were used and adopted from
Lestari et al. (
2022) and
Rincy (
2019) for the evaluation entrepreneurial awareness, five items were used and adopted from
Li and Wu (
2019) and
Rudhumbu et al. (
2016) for the assessment for indications of entrepreneurial education, and five items were adopted form
Rincy (
2019) and
Lestari et al. (
2022) for the evaluation of entrepreneurial intention.
The current study used partial least squares structural equation modeling (PLS-SEM 4) to estimate the proposed structure model suggested by
Hair et al. (
2021). The SEM-PLS model was investigated in two stages: model measurement and structural models. Convergent validity (content validity and reliability), factor loading, discriminant validity, and composite reliability were used to evaluate the accuracy of model measurement. The study used R-square, inner collinearity, and a level of significance of 5% was used for structural models.
Hair et al. (
2021) recommended that a construct needs to be both reliable and satisfactory if specific criteria are met, including a loading factor of 0.708, a composite reliability (CR) of 0.700, and a Cronbach’s Alpha value above the threshold of 0.700. The value of discriminant validity must be less than 0.850, and the extracted average variance (AVE) must exceed a threshold value of 0.500. In the next part, structural equation modeling will be used for creating hypotheses with a threshold level of significance of 5%.
3.5. Structural Model
After successfully measuring the measurement model, the next step is to calculate the structural model, which focuses on evaluating the relationships between the exogenous and endogenous constructs of the model. The first step in the structural model analysis is to evaluate the impact of the exogenous factors on the endogenous factors, utilizing the coefficient of determination (R-square) as a crucial step. According to the guidelines of
Chin et al. (
2010), the R-square value should range from 0 to 1, reflecting the proportion of variance in the endogenous variable that the exogenous variables can explain.
Therefore, as per the findings of R-square, it has been revealed that the coefficient of determination (R-square) for the constructs to entrepreneurial intention is 0.762, indicating that approximately 76% of the variance in the entrepreneurial intention variable can be attributed to the influence of exogenous factors such as entrepreneurial education, entrepreneurial skills, entrepreneurial awareness, and entrepreneurial inspiration. It also has proven the substantial contribution of these factors in shaping entrepreneurial intentions. Conversely, the remaining 24% of the variance is influenced by other factors not explored within this paper’s scope. This analysis underscores the significance of identified exogenous factors in fostering entrepreneurial intentions among university students. Second, the study shows that the R-squared value for entrepreneurial education is 0.257. This means that exogenous factors, such as entrepreneurial skills, entrepreneurial awareness, and entrepreneurial inspiration have 25.7% of the variance in entrepreneurial education (See
Table 6). The value of the R-square suggests that entrepreneurial education has a considerable influence. Understanding these dynamics can assist educators and policymakers in developing more effective measures to promote entrepreneurial education and, eventually, increase entrepreneurial intention among university students.
The second phase of the analysis involves employing structural equation modeling (SEM) to carefully test the proposed hypotheses, with a significance level of 5%. This step is crucial, as it allows researchers to assess the relationships between the constructs within the model after confirming that the model has been effectively measured and validated. The process of hypothesis testing means evaluating whether the exogenous factors significantly influence the endogenous variables, as recommended by
Hair et al. (
2021). In this study, seven precise hypotheses have been formulated, detailed in
Table 7. Each hypothesis hypothesizes a relationship between various exogenous factors, such as the impact of entrepreneurial inspiration, entrepreneurial skills, and entrepreneurial awareness on endogenous variables, including entrepreneurial education as a mediator and entrepreneurial intention.
In addition, the results of the hypotheses testing indicate that the exogenous factors (entrepreneurial inspiration, entrepreneurial skills, and entrepreneurial awareness) have a statistically significant impact on the endogenous variables (entrepreneurial intention). The finding reinforces that the identified exogenous factors play a critical role in influencing the outcomes of entrepreneurial education and entrepreneurial intention, thereby validating the study’s theoretical framework. The finding of the hypotheses is highlighted in
Table 7 at 5% of significant levels.
The analysis provided compelling evidence supporting Hypothesis 1, which posits a significant relationship between entrepreneurial inspiration and entrepreneurial intention among Pakistani university students. The statistical analysis revealed a
p-value of 0.001, which is considerably below the 5% significance level (alpha = 0.05), as recommended by
Hair et al. (
2021). This finding underscores a strong statistical significance between the two variables. Additionally, the t-value associated with this relationship was 3.002, exceeding the threshold of 1.645, further validating the hypothesis.
Hypothesis 2, which emphasizes a significant correlation between entrepreneurial awareness and entrepreneurial intention, was also supported by the statistical analysis. The analysis yielded a p-value of less than 0.015, again below the 5% significance level. The path coefficient of 0.277 indicates a positive relationship, suggesting that higher levels of entrepreneurial awareness are associated with a greater likelihood of individuals expressing entrepreneurial intentions. The t-value for this relationship was calculated to be 2.175, reinforcing the statistical significance of the hypothesis.
Regarding Hypothesis 3, the analysis speculated a significant relationship between entrepreneurial skills and entrepreneurial intention. The results showed a p-value of 0.011, which is well below the 5% significance level, indicating a strong statistical association between entrepreneurial skills and intention. The path coefficient was 0.252, implying that, as an individual’s entrepreneurial skills increase, so does their intention to engage in entrepreneurial activities. The t-value of 2.296, which exceeds the recommended threshold, further supports the hypothesis.
For Hypothesis 4, the analysis explored the relationship between entrepreneurial education and entrepreneurial intention, revealing a significant connection. The p-value was less than 0.009, below the 5% significance level, and the path coefficient was 0.269, quantifying the strength and direction of this relationship. The t-value associated with this relationship was 2.374, again exceeding the threshold, indicating that increased entrepreneurial education leads to heightened entrepreneurial intention.
Hypothesis 5 examined the mediating role of entrepreneurial education between entrepreneurial inspiration and intention. The statistical analysis yielded a p-value of less than 0.023, which is below the 5% significance level. The path coefficient of 0.545 and the t-value of 3.763, which exceed the threshold, suggest that entrepreneurial education significantly strengthens the relationship between inspiration and intention, highlighting its crucial mediating role.
Similarly, Hypothesis 6 investigated the mediating effect of entrepreneurial education on the relationship between entrepreneurial awareness and intention. The analysis showed a p-value of less than 0.002, significantly lower than the 5% significance level, indicating a strong mediation effect. The path coefficient was 0.520, and the t-value was 3.550, further confirming the importance of entrepreneurial education in enhancing the relationship between awareness and intention.
Lastly, Hypothesis 7 explored the mediating effect of entrepreneurial education on the relationship between entrepreneurial skills and intention. The analysis revealed a
p-value of 0.001, below the significance threshold, suggesting a statistically significant mediation effect. The path coefficient was 0.512, and the t-value was 2.955, indicating that entrepreneurial education plays a vital role in mediating and enhancing students’ entrepreneurial intentions, instilling confidence and empowering them to pursue entrepreneurial activities. A detailed results regarding structural model highlighted in the
Figure 3.