THEORETICAL FRAMEWORK

Similar to previous studies, the authors will design the research on popular theoretical foundations. Firstly, we employed the Theory of Entrepreneurial Event Model (EEM), which was first developed by Shapero & Sokol5. Specifically, this model assumed that entrepreneurial intention stems from individual desire, feasibility, and ability to seize opportunities6. Secondly, the study would apply the Model of Implementing Entrepreneurial Ideas, which was developed by Bird 7. This model suggested that individuals tend to form intentions "based on a combination of both personal and contextual factors" 8. Finally, the Theory of Planned Behavior (TPB) would be applied. It is supposed that behavior came from three factors: attitudes toward the behavior, subjective norms, and perceived behavioral control 9.

LITERATURE REVIEW

Many studies have been conducted to find the factors that affect students’ Entrepreneurial Intentions. For instance, a study by Shahverdi & Qureshi is one of several international publications conducted on this topic in Malaysia10. The authors used CFA and SEM to find out factors affecting entrepreneurial intention, in which, lack of competency, lack of self-confidence, and lack of resources are considered direct barriers; meanwhile, lack of support and knowledge is only considered indirect factors. Similarly, Annuar et al. also proved that personal traits, entrepreneurial skills, and micro level are the factors creating barriers11. In addition, Amanamah et al. also used a survey sample consisting of 731 respondents to find out the factors that hinder start-ups in Ghana12. By multivariate regression, the authors believed that economic factors had the strongest effect on entrepreneurial intention, in contrast, the personal factor has a rather weak influence. Through the research in Bosnia and Herzegovina, Turulja et al 13 showed that the support of family and friends exerted a significant positive influence on entrepreneurial intentions. Fear of failure had a significant adverse impact on entrepreneurial intentions while entrepreneurial capacity enhanced it. The study at the University of Mongolia by Zanabazar & Jigjiddorj 14 aimed to explore various factors affecting the entrepreneur intention, including attitude of the students, subjective norms, entrepreneur education, and perceived behavior control. More than 500 university students were involved in the data sample, in which the valid respondent had to attend the entrepreneurship subject. The survey results demonstrated that personal attitude had an influence on entrepreneurial intention and the participants expressed their willingness to start their businesses by having an awareness of prospective challenges and opportunities.

In Vietnam, many studies on the same topic have been conducted. Thu et al. 15 and Trang16 used the SEM model to find out the factors affecting the entrepreneurial activities of students. In particular, the study of Thu et al. 15 clarified the role of cognitive factors, while Trang 16 focused on supporting factors and barriers. By other methods, Mai et al. 17 and Lien 18 combined EFA with multivariable regression to explore what factors and how they influenced entrepreneurial intention. Based on these two studies, the factors could be listed as mental support, capital, education, and personal characteristics of students17, 18. On the same topic, Thanh et al.19 used logistic regression, combined with correlation coefficient and EFA to catch on barriers affecting entrepreneurial intention. The results from this study showed that personal traits, cognitive conditions, and normative and regulative structures limited the desire to start a business in Vietnamese students, with the strongest impact being Personal traits specifically. Hien & Trang20 based on the theory of intended behavior of Ajzen9 combined with related studies to build a proposed research model. However, the author ' research had some limitations as the study only surveyed final-year students and ignored the others. Besides, the independent variables in the research model only explained 55.1% of the variation of the dependent variable. This meant that although the research model was suitable, 44.9% still belong to other factors not mentioned in the model. On the other hand, Van, Y, and Ha 21 collected primary data from 250 economics students at Tra Vinh University (TVU). Thanks to multivariable regression analysis, the study found six factors affecting the start-up intention of economic students including: start-up support; feasibility perception; educational environment; personality traits; attitudes towards start-up behavior; financial accessibility. Hiep et al. 22 collected data from 430 final-year economics students from 10 universities that had the highest rate of start-up students in Ho Chi Minh area. After applying the Multiple Linear Regression Analysis Model, the research results showed that the factors affecting the intention to start a business of economics students at universities in Ho Chi Minh City (arranged in order of importance from high to low) include: business education; subjective standards; startup environment; personality characteristics and perception of feasibility.

From the review process, it can once again be affirmed that the previous studies all have certain gaps. This gap can be easily seen through the lack of barriers affecting entrepreneurial intention, the lack in the case of Economics and Management students, and the lack in the case of universities in Ho Chi Minh City. Thereby, the research reinforces the research orientation on barriers to entrepreneurial intention among Economics and Management students in Ho Chi Minh City.

CONCEPTUAL FRAMEWORK

The model of this study focuses on clarifying the barriers affecting the entrepreneurial intention of Economics and Management students, w herein entrepreneurial intention is a process from thinking ( plann ing actions) to carrying out an entrepreneurial behavior 23.

Specifically, the study proposes six barriers affecting business intention as follows:

The first is knowledge barriers. Lack of knowledge and skills is considered a serious barrier to business intention24. Indeed, Miller25 identified a lack of knowledge and business skills as another potential barrier to entrepreneurship intentions. Therefore, the authors proposed that the knowledge barrier factor had a negative impact on entrepreneurial intention of students.

The second is cognitive barriers. According to Thu et al15, perception included desirability and feasibility. While desirability refered to the perceived value and attractiveness of the opportunity, feasibility included the possibility of implementation and the constraints of the opportunity 26. Taatila 27 also concluded that the lack of such awareness might adversely affect one's choice to start their own business. Therefore, the authors propose that cognitive barriers have a negative impact on students' entrepreneurial intentions.

The third is market barriers. A market is a place where buyers and sellers exchange goods and services. External market factors can have a positive or negative impact on an individual's thinking, so they also cause the entrepreneurial intention to change28. Therefore, the authors propose that market barriers positively or negatively affect students' entrepreneurial intentions.

The fourth is mental barriers. Many previous studies listed “mental barriers” as the negative factor affecting entrepreneurial intention, such as the study by Bich & Minh 29 and Herdjiono et al 30. Indeed, when individuals had no mental support, they tended to become self-deprecating when making startup decisions. In summary, mental barriers are proposed to negatively impact on the intention to start a business.

The fifth is the capital barriers. Capital is essential for the survival of the business in the early stages 31. The lack of capital is considered one of the critical factors hindering the intention to start a business. Therefore, the authors propose that the capital barriers negatively impact entrepreneurial intention.

Finally, there are barriers relevant to the educational environment. In the university environment, students have the freedom to be creative and come up with their entrepreneurial ideas32. According to Lüthje and Franke 33, training programs and university career-oriented activities could increase students' interest and perceptions of entrepreneurship in the future. In addition, some studies also consider education as a basic requirement when analyzing entrepreneurial intentions. In particular, Hiep et al. 22 affirmed that the Educational Environment is the most important factor affecting the entrepreneurial intention of economic students. Therefore, the authors propose that the educational environment barriers have a negative impact on entrepreneurial intention.

Generally, these factors are presented as the following observed variables (Table 1).

On the other hand, to increase the significance of the topic, some demographic factors and personal characteristics were also included in the analysis17, 18. These demographic factors include (1) Gender, (2) Year of study, (3) School, (4) Field of study, and (5) Occupation of parents.

In sum, the conceptual framework could be illustrated in Figure 1, which was also the research model for this study.

H1: Entrepreneurial intentions are affected by knowledge barriers.

H2: Entrepreneurial intentions are affected by cognitive barriers.

H3: Entrepreneurial intentions are affected by market barriers.

H4: Entrepreneurial intentions are affected by mental barriers.

H5: Entrepreneurial intentions are affected by capital barriers.

H6: Entrepreneurial intentions are affected by educational environment barriers.

DATA

The study carried out primary data collection with 312 valid questionnaires, surveyed from February to March 2022 by convenience sampling technique in Ho Chi Minh City. This sampling method is appreciated for its efficiency, simplicity, and cost-saving 34. Additionally, to reduce the cost, the authors will only focus on some economics-teaching universities in HCMC, including UEL, UEH, UEF, and TDTU. These universities typically enroll a large number of economics-management students per year, which are expected to be representative of the students in the same major in HCMC35, 36, 37, 38.

T he surveyed demographic factors were (1) Gender: male and female; (2) School year: a maximum of 4 years; (3) Universities : UEL, UEF, UEH, and TDTU (4) Fields of study comprised of Real Estate business, Economics - Public Management, International Economics Relations, International Business, Business Administration, Economics, Marketing, Commerce and some other disciplines; and (5) Parents' occupation including non-business, business-related, public servants and self-employment.

In addition to demographic factors, the authors also built a questionnaire according to measurement variables representing barriers. In which, knowledge barriers (KT), and market barriers (MA) had 6 observed variables, cognitive barriers (NT), capital barriers (NV), educational environment barriers (GD) had 5 observed variables, and mental barriers (TT) had 4 observed variables. These observed variables were evaluated on a Likert scale from 1 to 5, with 1 being “Strongly Disagree” and 5 being “Strongly Agree”. Besides, the entrepreneurial intention was also built based on 5 observed variables including the questions: (1) I cannot start a business, (2) I cannot be self-employed in the future, (3) I am not thinking about starting my own company, (4) I have no goal of becoming an entrepreneur, and (5) I am not ready to learn how to start a business.

ANALYTICAL METHOD

From the collected data, the authors conducted data processing and analysis. Specifically, the study used SPSS 20 and AMOS 24.0 software to perform Cronbach's Alpha Test, Exploratory Factor Analysis (EFA), Confirming Factor Analysis (CFA), Structural Equation Modeling (SEM), and Kruskal – Wallis test for non-normally distributed data.

Moreover, to increase the relevance and reliability of the study, the authors also tested the dispersion and convergence for the CFA result through the CR, AVE, and Heterotrait-Monotrait Ratio of Correlations (HTMT) Indexes. Similarly, the authors also test Bootstrap (with 500 observations) for linear structural model SEM.

CRONBACH’S ALPHA RESULT

The results of Cronbach's Alpha test (Table 2) showed that all scales had acceptable Cronbach's Alpha coefficients (greater than 0.6). Specifically, the lowest factors were the cognitive barriers and the capital barriers (with the same Cronbach's Alpha equal to 0.755), and the highest one was the mental barriers scale (with Cronbach's Alpha equal to 0.878). Moreover, the item-total correlation for all observed variables was greater than 0.3, therefore, they could be included in the EFA and CFA exploratory factor analysis.

RESULT OF EXPLORATORY FACTOR ANALYSIS (EFA) AND CONFIRMATORY FACTOR ANALYSIS (CFA)

After 7 loops of performing Exploratory Factor Analysis (EFA), the number of remaining observed variables was 23 and the number of removed observed variables was 8 (its factor loading less than 0.35). In which, the KMO coefficient was 0.917 (greater than 0.5), the Bartlett Test was 0.0 (less than 0.05), and they were consistent with the above analytical assumptions. Moreover, all observed variables had a loading factor greater than 0.35, and the total variance extracted is 52.475% with the suitable eigenvalue.

After 3 loops of performing Confirmatory Factor Analysis (CFA), the number of remaining observed variables was 19 and divided into 4 factors. There was a new factor that differed from the original expectation, made up of observed variables TT1, TT3, NT1, TT2, and NT4 (Figure 2). Based on the content of the questions, this new variable could be defined as the “mental barriers” and the authors would use it for the next steps of analysis. The results of the CFA also showed appropriate indicators. Specifically, the index CMIN/df = 2,145; GFI = 0.905; CFI = 0.941; TLI = 0.931; RMSEA = 0.061; and PCLOSE = 0.03.

Besides, the authors also evaluated the dispersion and convergence of CFA. The results of Model Validity Measures represented that 4 factors KT, TT, GD, and MA had all CR values greater than 0.7 (acceptable for reliability) and all AVE values were greater than 0.5 (acceptable for convergence). Moreover, according to HTMT indexes, there was no correlation among factors, so the new barriers were guaranteed to be discriminant (Figure 2).

As a rule, these factors need to be calculated by Cronbach's Alpha again. Specifically, all new factors had Cronbach's Alpha values greater than 0.6, the smallest factor was the Market Barriers with a coefficient of 0.766 and the largest was the Educational Environment Barriers with a coefficient of 0.866.

In sum, after employing the EFA and CFA method , there were 4 constructed factors. Specifically, the EFA removed 8 observation variables, and the CFA removed the next 4 observation ones. Meanwhile, the rest were grouped as knowledge barriers, educational environment barriers, market barriers, and especially for mental barriers. And these variables would be used in the following steps.

RESULT OF STRUCTURAL EQUATION MODELING (SEM)

The SEM results (Figure 3) were consistent with the above expectations and suitable for the assumptions of the statistics. The following figure showed that CMIN/df = 2,290 < 3; CFI = 0.919; TLI = 0.908; RMSEA = 0.064; PCLOSE = 0.01, GFI=0.870 39, 40.

The estimated parameters (in Table 3) summarized that all factors were significant at 95%. In which, the factor mental barriers (TT) had the largest impact on entrepreneurial intentions (the coefficient was 0.562), then the market barriers (MA) (equal to 0.368), and the final one was education barriers (GD) (equal to 0.181). In contrast, the knowledge barriers (KT) had a negative effect on the intention (the coefficient was less than 0). Besides, the bootstrap coefficients with a 500-observation sample were significant at 95%, so the results of the SEM model were suitable for analysis.

RESULT OF KRUSKAL - WALLIS TEST

Entrepreneurial intention in this study was assumed to be a continuous variable and equal to the mean of the YDKN variables. Because this new variable did not follow a normal distribution, the authors decided to use Kruskal-Wallis Test to explore the association between demography and entrepreneurial intention.

Table 4 indicated that demographic factors influenced entrepreneurial intentions at a 95% significance level. Specifically, there were 5 groups with significant differences in the students’ entrepreneurial intention: (1) Gender, (2) Year of study, (3) School, (4) Field of study and (5) Occupation of parents. In sum, the above demographic factors had impacts on entrepreneurial intention, and provided scientific evidence for the last hypothesis.