Entrepreneurship has proven to be an effective tool for creating economic value and, at the same time, a means to address various social issues. Entrepreneurial intention is the inclination to start a new venture in the future. This intention indicates the potential of an entrepreneur to enter the pharmaceutical industry in the future. Krueger and Carsrud (1993) define entrepreneurial intention as the commitment to start a new business venture.

From a significant amount of previous research on entrepreneurial intention, three governing factors have been identified. The first is demographic factors, including age, gender, prior experience, and the influence of role models. The second is personality traits, including self-efficacy, self-confidence, autonomy, control ability, risk-taking propensity, and occupational attractiveness. The third factor is the context, including education and the environment (Wärneryd 1998). Some studies have examined the role of demographic variables in shaping students’ entrepreneurial tendencies, yielding different results (Wang and Wong 2004). On the other hand, Elali and Ai-Yacoub (2016) revealed that social networks have a significant impact on the development of entrepreneurial intention.

According to the theory of entrepreneurial characteristics, entrepreneurial intention is determined by specific characteristics. These include a high achievement need, defined as a willingness to accept financial risks, acceptance of ambiguity, referring to the lack of fear of the unknown, innovation, the ability to create or modify existing business concepts, and intuition, which involves planning for events before they occur (Nancy and George 1994). Additionally, these characteristics are the result of individuals’ experiences; they focus on opportunities for success and tend to avoid situations with low or high risk. This need motivates individuals to make a maximum effort to achieve their set goals. Alongside this, subjective norms also significantly influence entrepreneurial intention. Subjective norms involve an individual’s belief in oneself. Subjective norms are related to the social pressure that young pharmacists will face. Depending on the social environment, these pressures can either become motivators or barriers to entrepreneurial mental development. Financial pressures can act as a catalyst for individuals to seek new ways to generate income, especially if they are facing constraints in traditional employment or striving for financial independence. A study conducted in Viet Nam found that the intention of dispensing antibiotics without prescription of communtiy pharmacists might come from the pressure of customers and drugstore owners (Hung et al. 2023). Previous studies have concluded that subjective norms have a positive relationship with entrepreneurial intention (Kolvereid 1996a; Solesvik et al. 2012).

According to the environmental approach theory, entrepreneurial choices are related to external factors seen as cultural images, education, experience, and family background influencing individuals’ entrepreneurial intentions. Ahmed and Usman (2018) studied various factors to determine the determinants of students’ entrepreneurial intentions in Pakistan. They found that innovation and family business experience were related to entrepreneurial intention. Turker and Selcuk (2009) identified two significant factors (perceived educational support and perceived support structure) influencing university students’ intentions in Turkey. Bhandari (2006) studied the entrepreneurial intention of university students in India and found six factors: leadership of others, implementing innovative ideas, being my boss, determination, personal challenge, and non-business education are essential factors for starting a new business.

According to Ajzen (Ajzen 1991), the attitude towards entrepreneurial behavior reflects an individual’s positive or negative evaluation, support, or opposition to the intended action. It represents an individual’s favorable or unfavorable evaluation of the behavior. When pharmacists have a positive attitude towards entrepreneurship, perceive benefits, and have opportunities and resources, they will engage in entrepreneurship. Other factors influencing the attitude towards entrepreneurial behavior include risk acceptance, freedom, and independence (Krueger et al. 2000).

An insight from Bandura’s (Bandura 1986) social cognitive theory and Ajzen’s (Ajzen 1991) theory of planned behavior (TPB) is that, before engaging in behavior, individuals must have an intention and a plan for it. In the psychological behavior approach, intention is an important indicator influencing behavior when these behaviors are rare, difficult to observe, and occur at unpredictable times. According to Austin et al. (2006), entrepreneurship takes advantage of business opportunities to enrich oneself by initiating innovative activities in conditions of limited resources. Therefore, more broadly understood, entrepreneurship is an individual starting their own business activities (Begley and Wee-Liang 2001), whether an individual accepts the risk to start a new business by investing business capital or opening a business store (MacMillan and Katz 1992). In a narrower sense, entrepreneurship is a work attitude that emphasizes autonomy, creativity, innovation, and risk-taking to create new value for customers (Bird 1988).

According to the theory of reasoned action (TRA), behavioral intention is determined by individual attitudes and their perceived control. Subjective norms also influence attitudes by performing observable behaviors (Ajzen and Fishbein 1975). This theory was established to predict voluntary behavior and help individuals recognize the psychological factors of their behavior. The two main factors influencing intention are personal attitudes and subjective norms. Individual attitudes can be a positive or negative evaluation of an individual toward a behavior.

Intention is defined as motivating factors influencing someone’s behavior. Entrepreneurial intention can also be seeking information and other resources to start a venture (Katz and Gartner 1988), or it can be a commitment to undertake entrepreneurial behavior (Krueger et al. 2000). Entrepreneurial intention is the first step in the process of exploring, creating, exploiting opportunities for entrepreneurship, and starting new business activities (Gartner et al. 1994). Furthermore, previous studies have proven that knowledge is also a factor that promotes individuals toward entrepreneurship.

When researching the entrepreneurial intention of pharmacy students in Chennai, the hypotheses were placed in a favorable direction to entrepreneurial intention, including risk acceptance, psychological control point, achievement needs, job autonomy, entrepreneurial environment, and education (Sankar and Sudha 2016). Another study on entrepreneurial intention and internal business of pharmacy students in Iran researched factors related to achievement needs, leadership effectiveness, autonomy, significantly related to entrepreneurial intention. The relationship between having an entrepreneur in the family with entrepreneurial intention and the intrinsic motivation of entrepreneurs (nurtured in a family with a business foundation creates a special context that influences the future career intentions of children, family members can be a source of financial or non-financial support).

Perception of behavioral control (PBC): According to Ajzen (Ajzen 2002), perception of behavioral control can be seen as an individual’s perception of the difficulty of performing entrepreneurial behavior. Additionally, perception of behavioral control focuses on an individual’s belief in control over the behavior. That is, the degree of ease or difficulty in performing a behavior. The theory was first applied to explain entrepreneurial intention by Krueger and Carsrud (Krueger and Carsrud 1993), and since then, it has been demonstrated to reasonably explain entrepreneurial intention. Many previous studies have found empirical evidence showing that perception of behavioral control influences entrepreneurial intention because an individual with a high control ability will be more confident and will perceive higher feasibility, thereby encouraging individuals to engage in behavior. In this study, the authors argue that perceived control behavior significantly increases the likelihood of forming the entrepreneurial intention of students (Kolvereid 1996b; Tkachev and Kolvereid 1999). It also refers to a person’s perception of the personal feasibility of the mentioned behavior (Boissin 2005).

This study adopts a cross-sectional descriptive approach. Data were gathered through interviews with young pharmacists (aged 20 – under 40) using a pre-prepared questionnaire (either in print or Google form) focusing on startup knowledge from May 2023 to November 2023.

The study targets young pharmacists aged 20 to under 40, residing or temporarily living in various Mekong Delta provinces, who voluntarily participated in the research.

Young pharmacists capable of understanding the questionnaire and willing to participate in the study.

Young pharmacists providing incomplete survey responses.

Our research adhered to the guidelines outlined in the Declaration of Helsinki. The study protocol received approval from the Medical Ethics Council of Can Tho University of Medicine and Pharmacy, Vietnam (approval number 23.014.GV/PCT-HĐĐĐ). Prior to enrollment in the study, all participants provided written informed consent.

In investigating entrepreneurial intentions, the researchers employed a sociological survey method with a convenience sampling approach (Parket and Eilbirt 1975). Data were collected through a quantitative questionnaire administered through interviews with young pharmacists across 13 provinces in the Mekong Delta region, with a total sample size of 815.

To use Exploratory Factor Analysis (EFA), the sample size needs to be large and is determined based on the minimum sample size and the number of variables entered into the analysis. Following Carpenter’s recommendation of a minimum observation/variable ratio of 5:1, with 43 questions in the adjusted survey, the minimum sample size is 215. In this study, the actual sample size is 815, meeting the conditions for sample size (Carpenter S 2018).

The questionnaire consists of three parts:

Demographic Information: Including age, gender, workplace, living area, economic status, educational background, current occupation, and parents’ occupations and monthly income awareness of entrepreneurship.

Knowledge Survey: Comprising 14 questions, each scored based on correctness. Participants are categorized into three levels: Poor (<50% total score/below 7 points), Average (50%–70% total score/7 to 9.8 points), Good (>70% total score/9.8 points and above) (Mansour and Al-Kayali 2017).

Entrepreneurial Intentions: Focusing on the respondents’ perspectives on entrepreneurial intentions and influencing factors. All responses are measured on a 5-point Likert scale, ranging from “strongly disagree” (1) to “strongly agree” (5).

In summary, the study aims to investigate the impact of knowledge, attitudes, achievement needs, perception of behavioral control, and subjective norms on entrepreneurial intentions among young pharmacists in the Mekong Delta region from May to November 2023.

This model predicts and explains the tendency to perform behavior through the favorable influence of difficulty, attitude, achievement needs, perception of behavioral control, subjective norms on knowledge, and knowledge on entrepreneurial intentions (Fig. 1).

Figure 1. 

Research model.

Hypothesis H1: Difficulty has a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H2: Attitude has a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H3: Achievement needs have a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H4: Perceived behavioral control has a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H5: Subjective norms have a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H6: Knowledge has a positive impact on the entrepreneurial intentions of young pharmacists in the Mekong Delta.

This study employed a linear structural equation model and conducted various tests, including the reliability test of measurements using Cronbach’s Alpha, exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and structural equation modeling (SEM).

The Cronbach’s Alpha (CA) reliability coefficient was used to assess the reliability of measurements. This tool helped examine the reliability and goodness of the observed variables of the latent factor (Factor A). This test reflected the degree of tight correlation between observed variables within the same factor, indicating which variables contributed to measuring the concept of the latent factor and which did not. A good Cronbach’s Alpha result for the factor indicated that the listed observed variables were reliable, demonstrating the characteristics of the latent factor, and a good measurement scale for this factor was obtained.

Following this was the exploratory factor analysis (EFA). EFA was used to reduce a set of k observed variables into a set of F (F < k) more meaningful factors. The basis of this reduction relied on the linear relationship of factors with the primitive variables (observed variables). The extraction method used in factor analysis, specifically Principal Components Analysis with Varimax rotation, was the most common approach. The factor loading coefficients of variables had to be >= 0.5, and 0.5 < KMO < 1 indicated the appropriateness of EFA (Kaiser 1974).

Subsequently, confirmatory factor analysis (CFA) was conducted within the linear structural equation model, which has advantages over traditional methods such as exploratory factor analysis (EFA). This method improved the reliability and validity of measurements, enhanced the appropriateness of models, and facilitated more accurate conclusions and solutions. Finally, the structural equation model (SEM) was employed using AMOS to test the research questions and hypotheses in this study, a commonly used tool in recent scientific research. SEM tests relied on indices such as CFI, TLI, and these indices had to be greater than 0.9 (Bentler and Bonett 1980; Hair et al. 2011) to consider SEM as a statistically useful tool for testing research theories and conceptual models using experiential data. Moreover, SEM would determine the significant relationships between hypotheses in the research framework based on the collected data. Additionally, AMOS would be used to analyze qualitative data from interviews, which would be recorded and encoded according to predetermined themes.

Observational research conducted in the same interviews collected data on gender distribution. The female observation group accounted for 69.1%, while males comprised only 30.9%. There was a significant gender difference in the collected observations. Most participants were vocational/college school (46.5%) and worked at pharmacy/drugstore counter (42.6%). Most of their parents were farmers (41.7% and 48.3%) and had income > 5 million – 10 million (38.7%) (Table 1).

After describing the study sample, we proceed with the analysis of tests according to the research process. First, the study conducted Cronbach’s Alpha test for each research concept. Next, the study conducted Exploratory Factor Analysis (EFA) for the research sample, followed by Confirmatory Factor Analysis (CFA), and finally, Structural Equation Modeling (SEM) analysis to test the research model hypothesis on the entrepreneurial intentions of young pharmacists in the Mekong Delta.

After utilizing the SPSS software, the analysis of the Cronbach’s Alpha reliability coefficients for the independent and dependent variables showed reliability scores ranging from a minimum of 0.7 to a maximum of 0.95. A correlation greater than 0.3 indicates reliability. Thus, all variables within the scale met the requirements and were accepted for analysis in subsequent variables. When the Cronbach’s Alpha is greater than 0.95, the variables are almost equal to 1. The results of the Cronbach’s Alpha analysis in this study indicated that the scales of the 06 factors were all above 0.8, ensuring the reliability and validity requirements (O’Hair et al. 1998). Among them, variables NCTT3, A7, SN8, PBC8, and KK12 were excluded due to having higher current CA scores, and I5 was excluded because its score of 0.075 was below the minimum CA threshold (Table 2).

After removing inappropriate variables, the scales were subjected to Cronbach’s Alpha reliability analysis and then utilized in the Exploratory Factor Analysis (EFA). This method aims to assess the convergence and discrimination of the conceptual scales. EFA does not differentiate between independent and dependent variables but relies on the correlation between observed variables. Each observed variable is assigned a score called the factor loading.

Key indices in EFA include the Kaiser-Meyer-Olkin (KMO) coefficient, the number of factors extracted at Eigenvalues >= 1, factor loading coefficients, and the total explained variance. The KMO coefficient, with values sufficiently large (0.5 < KMO < 1), indicates that the conditions for factor analysis are appropriate.

In our study, all observed variables meet the requirements, with KMO coefficients ranging from a minimum of 0.5 (for achievement motivation analysis) (Table 3) to a maximum of 0.923 (for attitude analysis) (Table 4). This demonstrates statistical significance and confirms that the data are sufficiently suitable for exploratory factor analysis. The results also show compliance with the requirement for total variance explained (>50%). The EIGENVALUES coefficients are all greater than 1 (ranging from 1.841 to 5.046), with P-values of 0.000.

The analysis results meet the requirements with KMO = 0.923. Bartlett’s test yields a significance value (Sig) of 0.000, Eigenvalue = 5.001 (> 1), and one factor is extracted. The total explained variance is 83.348%, exceeding 50%, satisfying the conditions for EFA. The analysis is conducted with the observed variables in the independent component (A): Attitude.

The analysis results meet the requirements with KMO = 0.902. Bartlett’s test yields a significance value (Sig) of 0.000, Eigenvalue = 4.515 (> 1), and one factor is extracted. The total explained variance is 64.494%, exceeding 50%, satisfying the conditions for EFA. The analysis is conducted with the observed variables in the independent component (SN): Subjective Norm (Table 5).

The analysis results meet the requirements with KMO = 0.870. Bartlett’s test yields a significance value (Sig) of 0.000, Eigenvalue = 4.382 (> 1), and one factor is extracted. The total explained variance is 62.603%, exceeding 50%, satisfying the conditions for EFA. The analysis is conducted with the observed variables in the independent component (PBC): Perception of Behavioral Control (Table 6).

The analysis results meet the requirements with KMO = 0.934. Bartlett’s test yields a significance value (Sig) of 0.000, Eigenvalue = 6.121 (> 1), and one factor is extracted. The total explained variance is 55.648%, exceeding 50%, satisfying the conditions for EFA. The analysis is conducted with the observed variables in the independent component (C): Challenges (Table 7).

The analysis results meet the requirements with KMO = 0.784. Bartlett’s test yields a significance value (Sig) of 0.000, Eigenvalue = 2.870 (> 1), and one factor is extracted. The total explained variance is 71.762%, exceeding 50%, satisfying the conditions for EFA. The analysis is conducted with the observed variables in the independent component (I): Intention (Table 8).

To sum up, most of the concepts ensure both discriminant and convergent validity with the researched data.

The CFA results show that the model has TLI = 0.945, CFI = 0.954, GFI = 0.900 (> 0.9), RMSEA = 0.049 (< 0.08), and Chi-square/df = 2.943 (< 3) (Table 9), which are considered appropriate for real-world data (Hu and Bentler 1999). According to Steenkamp and Van Trijp (1991), this reflects the unidimensionality of the scales. The factor rotation matrix for the results with 37 observed variables converges into one factor group, and all observed variables have factor loading coefficients > 0.5, indicating unidimensionality (Anderson and Gerbing 1988). There is no correlation between the residuals of the observations, achieving unidimensionality. All P-values are < 0.05, so the correlation coefficients for each pair of concepts differ significantly from 1 at a 95% confidence level (Steenkamp and Van Trijp 1991).

The CFA results of the measurement model are illustrated in the figure below (Fig. 2), with the following outcomes: the model has a Chi-square of 1630.641 (p = 0.000), CMIN-df = 2.943. The TLI and CFI indices are 0.945 and 0.954, respectively, both exceeding 0.9, indicating a well-fitting model. If these coefficients are greater than 0.9, the model is considered acceptable. RSMEA is 0.049, which is less than 0.08. In conclusion, the model fits the research data well.

Figure 2. 

Description of Confirmatory Factor Analysis (CFA) results.

The concepts have AVE > MSV, and √AVE is greater than the correlation between the concepts, affirming their discriminant validity. Additionally, the measurement scales for the concepts achieve the necessary convergence and reliability with AVE values of 0.529 and 0.609 (> 0.5) and CR values of 0.817 and 0.884 (> 0.7). The research model is suitable for market data and is appropriate for further CB-SEM analysis (Table 10).

From the linear structural model with 565 degrees of freedom, Chi-square = 1630.641 (p=0.000). The goodness-of-fit index (GFI), ranging from 0 to 1, achieved a good value of 0.900. The root mean square error of approximation (RMSEA) with a value of 0.049 meets the requirement. Currently, reports allow RMSEA to be less than 0.08 with 95% confidence. The comparative fit index (CFI) compares well, with a value above 0.9 considered good. The obtained CFI result is 0.954, indicating a good fit. The Tucker-Lewis Index (TLI), with a value close to 1, indicates a higher fit. The TLI result of 0.945 is good (Fig. 3, Table 11).

Figure 3. 

SEM testing results.

The table below illustrates that the factors impact as follows: The effects of difficulties, attitudes, perceived control, subjective norms, and achievement motivation on knowledge, and knowledge positively impact entrepreneurial intentions, all with significance levels <0.05 (Table 12).

Hypothesis H1: Difficulties have a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H2: Attitude has a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H3: Achievement motivation has a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H4: Perceived control has a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H5: Subjective norms have a positive impact on the knowledge of young pharmacists in the Mekong Delta.

Hypothesis H6: Knowledge has a positive impact on the entrepreneurial intentions of young pharmacists in the Mekong Delta.

This is consistent with the theoretical framework presented in the research.

In summary, entrepreneurial intentions are influenced by the factors of difficulties, attitudes, achievement motivation, perceived control, subjective norms, and knowledge. These impacts are positive and statistically significant.

It is essential to increase the number of pharmacists by continuing to collect additional data from other provinces and cities in Vietnam to enhance the reliability of the data.

Expand the scope of the study for a more comprehensive view.

Identify key and ancillary factors to avoid confusion and gain a better understanding of the results.