Entrepreneurial intentions among MBA students

Abstract Increasing entrepreneurial activities in a country start with an intention that leads to increased innovative activities, wealth creation, industrialization, employment generation, economic growth, and development. This paper examines the effect of attitude towards entrepreneurship, subjective norm, locus of control, entrepreneurial self-efficacy, and environmental support on entrepreneurial intention of 159 MBA students from two private universities in Ghana. The study uses structural equation modelling (SEM) to analyze the data obtained from the participants. The results show that all the factors but entrepreneurial self-efficacy significantly affects students’ entrepreneurial intentions. The study proffers policy-makers with the opportunity to nurture entrepreneurship in students as a foundation for transforming the intent into practice to address the huge employment gaps in emerging economies.


PUBLIC INTEREST STATEMENT
Undoubtedly, entrepreneurship is the leading driver of development in local, regional and national economies. This study empirically tests Azjen's TPB-based variables in the context of private universities in Ghana and in the process shedding some light in an area which has barely produced literature. Almost all of our respondents fall within the category of the working class, hence there are practical implications for private, public and voluntary businesses and industries in their quest to instil the entrepreneurial spirit and proclivity in the workers to cause transformation and a turnaround in their respective institutions.

Introduction
Intensifying entrepreneurial activities in a country has the possibility of contributing to innovative activities, creating wealth, increasing competition, industrialization, employment generation, and economic growth (Paul & Shrivastava, 2015). For developing and emerging economies, entrepreneurial activities could alleviate unemployment especially the case of graduates. Youth unemployment is high in Ghana and quite pronounced among graduates. For example, the rate of graduate unemployment in Ghana rose from 14. 7% in 1987to 40% in 2011(Baah-Boateng, 2015Zakaria et al., 2014). The unemployment scenario is primarily due to lack of skills and entrepreneurial consciousness among the youth. Another reason for the high graduate unemployment situations is the heavy reliance on the government for employment into the public sector. In the words of Johnmark et al. (2016, p. 2), "today's realities indicate that there is no government of any country that can absolutely provide jobs to absorb all graduates from her tertiary institutions. This means that there is the need for a change in the mindset of graduates from the look for a job syndrome to create a job mentality in order to actualize their educational aspirations".
In spite of extant research in the area of entrepreneurial intention, data from emerging economies is still insufficient. The majority of research on entrepreneurship has focused on developed countries (Nabi & Linan, 2011). For example, studies on entrepreneurial intention focus on developed countries such as Spain (Camelo-Ordaz, Diánez-González, & Ruiz-Navarro, 2016;Espiritu-Olmos & Sastre-Castillo, 2015), Poland (Nowinski & Haddoub, 2019), USA (Hsu et al., 2017). Thus, research using data from emerging economies is lacking. What widens this research gap is the over-concentration on public universities to the neglect of private universities whose existence has bridged the access gap to university education tremendously. We address this context gap by focusing this study on two private universities in Ghana. The objective is to determine which factors influence the entrepreneurial intentions of MBA graduates. According to Abiodun and Oyejoke (2017) intention is seen as the best predictor of entrepreneurial behaviour and intentionality is grounded on cognitive psychology that attempts to explain or predict human behaviour. The main assumption guiding this paper is that we believe the university environment is a fertile ground for breeding future and sustainable entrepreneurial activities needed for economic growth and national development. Hence, there is the need to identify, stimulate, and sustain student entrepreneurial intentions because entrepreneurs are not only born, but are also made.
Motivations for this study stem from prior research that qualifications acquired in postgraduate education influence entrepreneurial prospects through the acquisition of employment-related skills (Greene & Saridakis, 2008). According to Prodan and Drnovsek (2010) entrepreneurial intention is essential for entrepreneurial behaviour in academia. Some private universities including those from emerging economies (such as Ghana) have taken up the challenge by offering business programmes. In Ghana, as at the year 2018, there were 81 private universities (Ofori-Atta, 2019). Private universities in Ghana were established among other things, to augment the enrollment deficit in the public universities.
This study purposively selects two private universities to investigate the entrepreneurial intention among the MBA students. In a modified model, we complement the theory of planned behaviour with locus of control and environmental support as framework for the study after consulting extant literature (Oguntimehin and Olaniran, 2017;Esfandiar et al., 2019;Maes et al., 2014;Nowinski & Haddoub, 2019;Puni et al., 2018;Salami, 2019). We follow the recommendations of researchers in the field of psychology (Read et al. 2013;Yazdanpanah & Farouzani, 2015) who propose the inclusion of supplementary constructs to advance the predictive power of the TPB models. The reasoning for the adoption of the Theory of Planned Behaviour is based on its ability to explain human attitude towards a behaviour. Besides, entrepreneurship is a planned behaviour and cannot be created without sufficient planning (Jena, 2020). According to Ajzen (1991) the intentions to perform behaviours of different kinds can be predicted with high accuracy from attitudes toward the behaviour, subjective norms and perceived behavioural control (entrepreneurial self-efficacy). Some researchers, in predicting entrepreneurial intentions have replaced perceived behavioural control with entrepreneurial self-efficacy (Krueger et al., 2000;Miao et al., 2016;Moriano et al., 2012). Our study, therefore, focuses on entrepreneurial selfefficacy instead of perceived behavioural control. Two other variables (Environmental Support and LOC) that influence entrepreneurial intentions were incorporated in the model. Locus of Control illustrates a closer effect on people's intention to act (Esfandiar et al., 2019;Espiritu-Olmos & Sastre-Castillo, 2015). According to  environmental support profoundly influences entrepreneurial intention.
From a structural equation modelling (SEM) approach, the paper reports that, locus of control, attitude towards entrepreneurship, environmental support and subjective norm significantly affect student entrepreneurial mentality. By way of contribution, the study provides a framework for addressing graduate unemployment through detection of students with an entrepreneurial mentality. According to Obschonka et al. (2012) entrepreneurship seems to be the way of coping with massive unemployment and its attendant social vices among the youth. The study reinforces the Ghana government's efforts to address youth unemployment through various entrepreneurialbased flagship policies. Some of these interventions and institutions include Ghana Youth Employment Development (GYEEDA), National Youth Employment Programmme (NYEP), Youth Employment Agency (YEA), Youth Entrepreneurship Support (YES), National Entrepreneurship and Innovation Programme (NEIP). At the core of these programs is to nurture the entrepreneurial skills of the young Ghanaian by providing seed money and technical support to start new ventures.
The forgoing deliberations buttress the need to study the entrepreneurial intentions among MBA students among private university students. But unfortunately, no such rigorous study has been carried out among private university students to measure their entrepreneurial intentions. We believe this study may bridge the gaps in the literature and practice.
The organization of this article is as follows. The next section covers literature review and hypothesis development, a theoretical model to depict the various variables influencing entrepreneurial intention. The next section entails literature review and hypothesis development, research methodology, followed by results and discussion and lastly we draw the main conclusions and outline the implications and limitations of our research.

Literature review and hypothesis development
Many authors (e.g., Autio et al., 2001;Buli & Yesuf, 2015;Carayannis et al., 2003;Esfandiar et al., 2019;Hisrich & Peters, 2002;Kickul & Gundry, 2002;Krueger et al., 2000;Maes et al., 2014;Nowinski & Haddoub, 2019;Salami, 2019) have examined factors influencing entrepreneurial intentions. Against this backdrop, we have reviewed literature on some of the popular entrepreneurial intentions, using the Theory of Planned Behaviour as the base, to study as part of this paper. Ajzen (2019) defines intention as "a person's readiness to perform a given behavior". Ajzen (1991) posits that intention is the immediate determinant of behavior, professing that, "… the stronger the intention to engage in (planned) behavior, the more likely should be its performance" (p. 181). Bird (1988) indicates that entrepreneurial intention is a state of an individual mind, which directs and guides them towards the development and the implementation of new business concept. Van Gelderen et al. (2008) highlight entrepreneurial intention as the intentions of setting up one's business in the future. Prior research has established that entrepreneurial intent is the primary predictor of future entrepreneurs (Krueger et al., 2000). Krueger et al. (2000) suggest that entrepreneurial activity can be predicted more accurately by studying intention rather than personality traits or situational factors. Among the intention-based theories like the Theory of Entrepreneurial Event, Institutional Economic Theory and Theory of Planned Behaviour, the latter has more analytical capability (Diaz-Casero et al., 2012). The TPB is the most popular theory to explain the antecedent and consequences of entrepreneurial intention (Iakovlera, Kolvereid & Stephen, 2011).

Entrepreneurial intention
Besides, intention-based models contend that entrepreneurial venture creation must be preceded by the development of intentions to establish a start-up and by appreciating intentions we may be in better position to predict venture creation. Ajzen (1991) proposes that the intentions to perform behaviours of different kinds can be predicted with high accuracy from Attitudes Toward the behaviour, subjective norms and perceived behavioural control (entrepreneurial self-efficacy). Some researchers, in predicting entrepreneurial intentions have replaced perceived behavioural control with entrepreneurial selfefficacy (Krueger et al., 2000;Miao et al., 2016;Moriano et al., 2012). Our study, therefore, focuses on entrepreneurial self-efficacy instead of perceived behavioural control. Two other variables (Environmental Support and LOC) that influence entrepreneurial intentions were incorporated in the model (Esfandiar et al., 2019;.

Attitude towards entrepreneurship and entrepreneurial intention
Attitude Towards Entrepreneurship (ATE) refers to the degree to which one holds a positive or negative personal valuation about being an entrepreneur (Ajzen, 2001;Autio et al., 2013;Darren Lee-Ross, 2017;Krueger et al., 2000;Ozaralli & Rivenburgh, 2016). Previous studies by Autio et al. (2001) and Schwarz et al. (2009) have revealed that ATE was a major determinant in entrepreneurial intentions among respondents. Moriano et al. (2012) asserted that a positive attitude towards entrepreneurship was the strongest antecedent of entrepreneurial intentions. Thus, ATE is dominant in determining one's success or failure to overcome challenges when faced with equivocal situations in life (Darren Lee-Ross, 2017). Aragon-Sanchez et al. (2017) argued that an individual with a more positive attitude towards a given situation (e.g., entrepreneurial intention) is more likely to succeed as a person. Luthje and Frank (2003) observed that attitude toward entrepreneurship was the most important determinant of the intention to become self-employed and this attitude is influenced by the personality of the respondents. A number of authors (Aragon-Sanchez et al., 2017;Fini et al., 2012;Moriano et al., 2012) have established that the relationship between attitude towards entrepreneurship and entrepreneurial intentions is significant and the linkage has been proved in different circumstances. For instance, the small business founder's attitude towards entrepreneurial behaviour has been established to be the major factor of corporate entrepreneurial behaviour (Fini et al., 2012). Armitage and Conner (2001) maintained that there is a positive relationship between attitude towards entrepreneurship and entrepreneurial intention. Ayalew and Zeleke (2018), in studying the entrepreneurial intentions among engineering students in Ethiopia also found that ATE has a positive influence on students' self-employment. However, Gultom et al. (2020) established that attitude insignificantly influences intention among citizens of Indonesia, which was consistent with a paper by Zahid and Haji Din (2019).
From the forgoing, we hypothesize that: H 1 : Attitude Towards Entrepreneurship (ATE) has significant positive influence on entrepreneurial intentions.

Subjective norm (SN) and entrepreneurial intention
Subjective Norm (SN) is the perceived social pressure to perform or not to perform an entrepreneurial behaviour (Ajzen, 2001). Aragon-Sanchez et al. (2017) define SN as how an individual would behave in a particular setting. Entrepreneurship is associated with numerous changes and risks which may not be easily welcomed in an individual's lifestyle. This type of pressure could emanate from family members or the generality of society which forces an individual to do or not execute specific tasks. Hussain (2018) professed that a person would not desire to deviate from the norm and value held by close family members and even friends with whom one interacts on a regularly basis. The subjective norm has been perceived as traditionally weak, with respect to its role in the pattern of relationships in the TPB model, though this alleged weakness is not so clear. Nevertheless, some studies have simply omitted SN (Peterman & Kennedy, 2003;Veciana et al., 2005), while others found it to be non-significant (Autio et al., 2001;Krueger et al., 2000). Wijerathna (2015) showed that subjective norms and attitudes are the greatest factors that influence entrepreneurial intentions among agricultural students in Sri Lanka. However, Kankam and Abukari (2020) in their research in the eastern region of Ghana noted that attitude and subjective norms seem better predictors of intention than PBC. Linan and Chen (2009) report that, in the specific area of entrepreneurship research, only 7 out of the 16 studies previously reported included SNs in the analysis but two of them did not perform any regression analysis. Of the remaining five studies, three found SN to significantly explain EI (Kolvereid, 1996b;Kolvereid & Isaksen, 2006;Tkachev & Kolvereid, 1999), whereas the other two found SN to be nonsignificant (Autio et al., 2001;Krueger et al., 2000). Therefore, although there is support for the idea that a direct SN-EI relationship might be established, some controversy remains. Kuada (2015) professed that certain cultural traits have the tendency of influencing entrepreneurial intentions. Ghana is a highly collectivist country (Hofstede, 2012) and Gelaidan and Abdullateef (2017) suggest that relation support (e.g. emotional support or access to startup capital from family and friends) is a fundamental ingredient in nurturing entrepreneurial intentions in people. However, in highly collectivist cultures, people's inclination to become entrepreneurs is less (Autio et al., 2013;Takyi-Asiedu, 1993). Despite this assertion, collectivist values are necessary in nurturing entrepreneurship through the utilization of the requisite business resources (Lechler, 2001;Tiessen, 1997), the promotion of consumers' acceptance of entrepreneurs' innovations (Rauch et al. (2013). For instance, support from family and friends would boost people's confidence in the engagement of entrepreneurial activities, but its absence would serve as a disincentive. Gelaidan and Abdullateef (2017) in examining the entrepreneurial intentions in Malaysia argue that role models, family members and friends can provide economic and emotional support to the prospective entrepreneur. Gultom et al. (2020) in a study in Indonesia found that subjective norms have significant and positive influence on intention.
Therefore, we hypothesize that: H 2 . Subjective norm significantly and positively influences entrepreneurial intention.

Locus of control and entrepreneurial intentions
According to Rotter (1990) Locus of control (LOC) is the ability of individuals to control the events in life. Locus of control depicts the perception of one's ability to influence the outcome of a behaviour (Hsiao et al., 2016). Locus of Control has proven its importance in affecting the level of aspiration for entrepreneurship (Luthje & Frank, 2003;Rauch & Frese, 2007b). The Locus of Control theory has two categories of control perceptions (Ng et al., 2006;Zigarmi et al., 2018); internal or external and each has a differential influence on entrepreneurial intention. Bönte and Jarosch (2011) explored that an individual with a higher internal locus control is inclined towards self-employment because they are optimistic that their destiny is in their own hands. However, with an external locus of control, their life is determined by external circumstances like chance, luck or fate. Khan, Ahmed, Nawaz and Ramzan in a study in 2011 indicated that students with internal locus of control will display a positive inclination towards entrepreneurial intention. Earlier accounts on internal locus of control and entrepreneurial intention produced inconsistent and conflicting results Gurol & Atsan, 2006;Rauch & Frese, 2007). Prior studies have revealed that students with higher internal locus of control are high in entrepreneurial behaviour and entrepreneurial intention (Gurol & Atsan, 2006;Koh, 1996;Mazzarol et al., 1999;Mueller & Thomas, 2001;Vodă & Nelu, 2019). However, Ferreira et al. (2012) and Dinis et al. (2013) did not register any significant correlation with entrepreneurial intentions. Chaudhary (2017) in a study of India University students observed that successful entrepreneurs have an internal locus of control compared to ordinary people. From the forgoing, and in the wake of these contradictions, we expect that people with an internal locus of control to have a positive inclination towards entrepreneurial career (Ajzen, 1991;Esfandiar et al., 2019).

Entrepreneurial self-efficacy (ESE) and entrepreneurial intention
Self-efficacy is "an individual's belief in one's capacity to organize and execute courses of action required to produce given attainments" (Bandura, 1997, p. 3). Chen et al. (1998) define entrepreneurial self-efficacy as the strength of an individual's belief that he or she is capable of successfully performing the various roles and tasks of entrepreneurship. Entrepreneurial self-efficacy "measures a person's belief in their ability to successfully launch an entrepreneurial venture" (McGee et al., 2009, p. 965) and calls for success in activities like innovation, marketing, management and finance which are relevant to the creation of an entrepreneurial venture (Chen et al., 1998;Hsu et al., 2017). Entrepreneurial self-efficacy is an important antecedent of entrepreneurial intention (Krueger et al., 2000;Salami, 2019).  argued that there is a significant positive relationship between ESE and entrepreneurial intentions of students and working people alike.
Subsequent to the emergence of Ajzen's Theory of Planned Behaviour, a crucial line of research emerged to assess the link between ESE and entrepreneurial intention (Engle et al., 2010;Hsu et al., 2019;Kickul et al., 2009). This is probably due to the fact that empirical studies shown a significant positive relationship between ESE and Entrepreneurial Intentions (Barbosa et al., 2007;Chen et al., 1998). Prodan and Drnovsek (2010) emphasized that self-efficacy is the most significant variable in the explanation of academics' entrepreneurial intentions as compared with other predictors. In conceptual terms, there is no difference between perceived behavioural control and self-efficacy (Ajzen, 2019). Authors like Ajzen (1991), Schwarz et al. (2009), andTrivedi (2016) perceive that PBC and Entrepreneurial Self-efficacy constructs as interchangeable. But, Terry (1993) has proposed that ESE and PBC are not entirely synonymous. For instance, Bandura (1992) has argued that PBC and ESE are quite dissimilar concepts. That is Self-efficacy is more concerned with cognitive perceptions of control based on internal control factors whereas PBC is more generally an external factor. Bandura (1997) sees Self-efficacy as close to the "perceived behavioral control" in Ajzen's model. Perceived Behavioural Control bears a resemblance to the Theory of Perceived Self-Efficacy (Moriano et al., 2012) and for this study, PBC is substituted with ESE. This is not uncommon since some researchers (Hockerts, 2017;Tran & Von Korflesch, 2016) have applied one or more of the exogenous constructs in the TPB-based model. Armitage and Conner (2001) see Self-efficacy as a stronger predictor of entrepreneurial intentions. There is overwhelming empirical evidence to support a positive relationship between ESE and Entrepreneurial Intentions (Aragon-Sanchez et al., 2017;Chen et al., 1998;Krueger et al., 2000;Luthje & Frank, 2003). Some researchers (Krueger & Brazeal, 1994;Luthje & Frank, 2003;Pittaway et al., 2010) have opined that, the greatest the belief that the individual has in their abilities, the greater the entrepreneurial intention. Gielnik, Bledow and Stark (2019)'s paper on Tanzanian and Rwandan students showed that variability and the average in entrepreneurial self-efficacy participants displayed during an entrepreneurial training were positively related to business ownership in the succeeding year. In line with the preposition that self-efficacy helps people to generate the motivation to enhance their intentions, entrepreneurial research has found that nascent entrepreneurs are more likely to start and successfully manage a business when their entrepreneurial self-efficacy is high.
Thus, we posit that: H 4 : Entrepreneurial Self-efficacy is positively related to entrepreneurial intention

Entrepreneurial intention and Environment Support
The Eurobarometer Survey on Entrepreneurship reports that a lack of business experience, the challenge of raising start-up capital, red tape, the poor economic environment and an innate fear of failure were to be blamed for inhibiting more of Europe's potential entrepreneurs from venturing into entrepreneurship. Stephen et al. (2005) point out government support measures and processes as fundamental in the decision to start a firm. According to Van de Ven (1993), entrepreneurial research without reference to the environment should be considered as insufficient and incomplete. Environmental forces can be a major inhibitor to the creation of an entrepreneurial venture. Prior studies have revealed that significant environmental antecedents of entrepreneurial intentions include access to capital (Luthje & Frank, 2003;Ozen Kutanis et al., 2006;Schwarz et al., 2009), knowledge of potential business sector (Kristiansen & Indarti, 2004) and social networks (Sequeira et al., 2007). Luthje and Frank (2003) emphasized that a student might be prepared to establish a company, notwithstanding his relatively bad inclination towards entrepreneurship, because he perceives the founding conditions as very favourable. On the other hand, graduates with a positive attitude towards entrepreneurship may not decide to venture into their own business due to negative perception of critical factors in the environment. Thus, one of the fundamental challenges facing students with an entrepreneurial intention and activities is lack of enabling and supportive environment (Indarti, Rostiani & Nastiti, 2010;F. Khan et al., 2014). These scholars assert that the correlation between environment and entrepreneurial intention is worth researching into. Access to capital, as a variable of the environment, is undoubtedly one of the important determinants in establishing a new business (Kim et al., 2006;Kristiansen & Indarti, 2004). A considerable numbers of people have forsaken their nascent entrepreneurial careers because of an inability to access capital (Marsden, 1992;Meier & Pilgrim, 1994) and in Ghana the situation is even pathetic due to, among other causes the high-interest rates financial institutions charge for loan acquisition. Start-up capital can be procured from personal savings, family, friends, and bank loan or via partnership with an investor (Cetindamar et al., 2012). Prior studies in some developing countries propose that the availability of institutional support enhances growth of entrepreneurial firms (Amankwah-Amoah & Debrah, 2017; Donbesuur et al., 2020;Nakku et al., 2020). Urbano et al. (2020) in a sample of 14 developing countries came out that access to bank credit has a positive effect on entrepreneurship in developing countries. However, Ge et al., 2017) suggest that institutional support might not necessarily lead to successful entrepreneurial outcomes. Cetindamar et al. (2012) have emphasized that regardless of gender, financial capital is a crucial force for any subsequent entrepreneurial activities. Indarti et al. (2010 andYar et al., 2008) found in their study that environment is a significant factor in influencing entrepreneurial intentions.
From the abovementioned empirical discussion and evidence from prior studies of various researchers in entrepreneurial intention and environment support, the succeeding hypothesis is framed as follows: H 5 : Environment support is positively related to entrepreneurial intention

Research methodology
The research design is exploratory research, where the researchers focused on investigating and examining factors influencing students' entrepreneurial intention. We adopted a quantitative research approach to measure constructs, model the relationships between the variables. The data collection technique is the use of questionnaire. The respondents completed a set of seven items influencing entrepreneurial intentions. With the exception of the demographic characteristics, the entire responses format was a 5-point Likert-type scale. The items included entrepreneurial self-efficacy, perceived behavioural control, subjective norm, attitude towards entrepreneurship, locus of control, environmental support and risk-taking propensity. The seven items constituted the independent variables and the dependent variable was entrepreneurial intention. However, the principal component analysis reduced the independent variables items to five.

Measures
This research adopted items from previous TBP-based studies (e.g., Chen et al., 1998;Greene & Rice, 2007;Krueger et al., 2000) to measure the constructs due to their established construct reliability and validity, as well as their relevance to the purposes of this study. In exploratory studies, values ranging from 0.60 to 0.70 are considered acceptable (Hair et al., 2017). But according to Feldt and Kim (2008), a cut-off value of 0.70 is recommended. We followed Eddleston and Powell (2012) and Powell and Eddleston (2013) and performed confirmatory factor analysis (CFA), which provides a more rigorous test of validity (Cheung & Lau, 2008)). The application of a multi-item scale is highly recommended over less reliable single-item measures (Armitage & Conner, 2001); hence, the variables had more than one item. The instruments used for soliciting information from the participants are described in the following section.  Table 1.
Attitude towards Entrepreneurship was measured with an adapted questionnaire by Kolvereid (1996). The Cronbach Alpha value for Attitude towards Entrepreneurship is 0.720 as depicted on Table 1, compared to Kolvereid's (1996) values which ranged from 0.68 to 0.90, though he used a 7-point Likert-type scale.
Entrepreneurial Self-efficacy was measured with items from Wilson et al. (2007). The respondents were asked to rate their capabilities against their peers (1 = much worse, 5 = much better) in regards to solving problems, managing money, being creative, getting people's agreement, being a leader, and making decisions. Wilson et al. (2007) reported a Cronbach's alpha of 0.79. For this study, the scale yielded a Cronbach's alpha of 0.785. It is relevant to state that Perceived Behavioural Control was removed from the model after the running of the Principal Component Analysis. Thus, PBC was dropped because of insignificant contribution to prediction of intentions and because of problems with estimation (low reliability). Other empirical studies have also been unable to test this variable (e.g., Kolvereid & Isaksen, 2006;Lortie & Castogiovanni, 2015;Simon & Kim, 2017).
Locus of Control was measured with a ten-item developed by Mueller and Thomas (2001) and some of the items were reverse-coded. A sample of the items are 'when I get what I want, it is usually because I am lucky (Internal Locus of Control) and "success in business is mostly a matter of luck" (External Locus of Control). The Cronbach Alpha value for Locus of Control is 0.843 as depicted on Table 1.
Subjective Norm was measured with previous research by Kolvereid (1996) Table 1.
To measure Environmental Support, we adopted scales from Autio et al. (1997) on a five-point Likert-scale (1 = strongly disagree to 5 = strongly agree). The Cronbach Alpha value for Environmental Support is 0.803 as depicted on Table 1.
Entrepreneurial intentions served as the dependent variable and ATE, SN, ESE, LOC and ES were the independent variables. Prior studies by Zahid and Haji Din (2019) and Dalle et al. (2020) have reported and empirically investigated intentions as a dependent variable. Following prior studies (e.g., Oguntimehin & Olaniran, 2017;Esfandiar et al., 2019;Jena, 2020;Maes et al., 2014;Nowinski & Haddoub, 2019;Puni et al., 2018;Salami, 2019) we applied Likert-scale for the dependent and independent variables. Likert scale is used to measure psychological attitude, perception or opinion in a mathematical manner. This provides a more objective approach in measuring constructs, hence its choice in this research.

Sample and data collection
The population of this study was the students of Valley View University (the first private university in Ghana to charter, Techiman campus (VVU-TC) in the Brong Ahafo region of Ghana and Catholic University College of Ghana (CUCG), Sunyani in the Brong Ahafo region of Ghana. CUCG is affiliated to the Ghana's premier university (University of Ghana). The population of VVU-TC MBA students was 126 whereas that of CUCG was 76 as at the period of data collection. The respondents were Master of Business Administration (MBA) students with specialization in Banking and Finance, Strategic Management, Human Resource Management and Accounting. According to Krueger et al. (2000) and Shinnar et al. (2012) a student sample (e.g., master level students) is appropriate to study entrepreneurial intentions since students face immediate career choices and selecting an entrepreneurial career path is a viable alternative. Besides, relying on a student sample provides variation in terms of entrepreneurial intentions and attitudes (Shinnar et al., 2012). Thus, some students' entrepreneurial propensity will be positive whereas others will be negative.
The sample size was 159 out of the total 202 MBA student population from the two universities representing about 79% response rate. The study adopts a simple random sampling technique thus giving every student an equal chance of being selected. Data for the present study were collected via a self-administered questionnaire. The questionnaire was divided into eight sections. The first section (demographic variables) had eight questions, i.e., gender, age, level, marital status, employment status, sector, programme and educational background of respondent' parents. The response rate was 96% because the respondents answered the questions there and then (in their various lecture halls). Before administering the questionnaire to the students, they were briefed on the survey's objectives. To avoid bias in the responses, the students were assured of anonymity and confidentiality. Variables measured on a 1 to 5 scale with strongly disagree to strongly agree were used with respect to entrepreneurial intention.
Initially, seven thematic areas of entrepreneurial intention were developed in the questionnaire. In order to reduce these variables, we performed a principal component analysis to extract uncorrelated factors for further analysis (Cohen et al., 2003) and to validate the scale after the data collection (Saunders et al., 2009). However, before we performed the principal component analysis, we recoded some items (e.g., LOC) to obtain an empirical summary of the data set (Pallant, 2010). The PCA reduced the seven thematic areas to five.

Method of data analysis
The data collected were analysed using SPSS and Structural Equation Modelling (SEM) AMOS 7.0. To test the hypotheses and the proposed conceptual model, we used structural equation modelling (SEM), a tool that provides the appropriate and most efficient estimation technique for a series of separate multiple regression equations estimated simultaneously (Hair et al. (2014)). These authors posit that SEM is an appropriate technique for our study because it enables the usage of multi-item latent variables for an independent or dependent variable. SEM also has superior advantage of addressing measurement errors prevalent in such studies. There was no attrition, no missing data in all the variables. There were no outliers.

Model specification
The study aims to examine the factors that influence student entrepreneurial intention, using TPB as the foundation. We formulate a general regression model for entrepreneurial intention.
where Z i represents the dependent variable X 1 … X n are sets of explanatory variables and β 0 . . . β n are parameters to be estimated We introduce our variables into the general model as: Due to the complex nature of the constructs, we used several questions in order to avoid the danger of not covering key concepts of the variables. We employed factor analysis to reduce the number of items into their uncorrelated structures.

Results
The analysis of results covers a description of the demographic characteristics of respondents, factor analysis to reduce the number of constructs to their unobserved structures and analysis of the structural model showing which factors influence student entrepreneurial intention.

The current situation of MBA Student's entrepreneurial intentions
This research used two items (I love to create something different & I am determined to have my own business in the future) to measure Entrepreneurial intentions. The respondents' responses were divided into 5 grades (5 =strongly agree, 4 =agree, 3 =neither agree nor disagree, 2 =disagree, 1 =strongly disagree). On the item, I love to create something different, 47.8% and 40.3% went for Agree and Strongly Agree respectively. On the item, I am determined to have my own business in the future 32.1% and 61.0% chose Agree and Strongly Agree respectively. From these two items, we can say that the percentage of students with entrepreneurial intention is relatively high.

Demographic characteristics of respondents
The respondents were sampled from two private universities in the Brong Ahafo region of Ghana. Approximately 63% of the respondents were sampled from Valley View University-Techiman campus and the remaining 37% were from the Catholic University College of Ghana. Approximately 70% of the respondents were males and the remaining were females. This major disparity in gender at that level of education is not uncommon in Ghana. The highest age category was 30-39, representing 43% whilst the lowest stood at 3% for respondents who were 50 years and over. Approximately 65% of the respondents are married, whereas 35% were single. Approximately 90% of the respondents were Christians. This is understandable because in Ghana, Christianity is the dominant religion in Ghana. Ninety-four percent of the respondents are employed. This is justifiable because one of the entry requirements for most MBA programs in Ghana is work experience. That is, an applicant may be refused admission due to lack of work experience. Approximately 47% of the respondents were in Administration and Managerial positions and 60% were in the public sector. A significant percentage-33% of the respondents' parents had no formal education.

Factor analysis
Following previous studies (Hoque & Awang, 2016;Hoque et al., 2017;Nguyen et al., 2019), we employed exploratory factor analysis using principal component factoring to reduce the number of questions suitable for the model. The dimension reduction resulted in two items for attitude towards entrepreneurship and three questions each for the other constructs. The exploratory factor analysis uses the rotated component matrix for varimax to select variables with loadings above 0.7. The results show five components with Eigen values above 1 which explains total variance of 72.81%. We performed reliability analysis for the selected items to check internal consistency of the constructs using the Cronbach alpha values and the results show that all the constructs meet the threshold criteria of above 0.7 (Nunnally & Bernstein, 1994). The results can be seen in Table 1. Table 1 shows the construct measurement in the exploratory factor analysis covering the five variables used in the model. Apart from attitude towards entrepreneurship which was measured by two questions, all other variables have three questions. Using principal components extraction approach of factor analysis, the items were reduced to five uncorrelated components. The variances explained by each factor can be seen in the second column with a cumulative variance of 72.81%. The results for the measurement of internal consistency can be seen under the reliability column. Each variable has scores above 0.7, therefore indicating good constructs. The factor loadings from the rotated components (Varimax rotation method) shows that each item is above 0.7 Having reduced the dimensions of the constructs by exploratory factor analysis, we developed the model for predicting entrepreneurial intention among MBA students in emerging economies. Expanding the equation to include all the variables in the model, we have; where ATE4 = Starting a business will provide me with Independence ATE5 = Starting a business will provide me with opportunity to be my own boss

Structural model
We measured the structural model using structural equation modelling (SEM) with the AMOS software. For the path model, we employed the maximum likelihood estimation technique for SEM-AMOS to generate the coefficients for the measured and latent variables. We performed several goodness of fit analyses to ensure confidence in the structural model.

Measurement of goodness of fit
The formulation of the model was developed using the AMOS 16.0 software package. This analytical technique permits the evaluation of the overall fit of the proposed model and estimation of all corresponding coefficients simultaneously (Hair et al., 2017).
In order to check the absolute model fit, the root-mean-square error of approximation (RMSEA) and the goodness of fit index (GFI) were checked. The results show that GFI (0.932) is greater than 0.9 whilst RMSEA (0.029) is within the acceptable range. Ideally, acceptable RMSEA which indicates a good fit should be <0.08. Our results indicate a good model with root mean square error of approximation (RMSEA) less than 0.08 (Byrne, 2010). We performed further model fit analyses to check incremental model fit and parsimonious model fit. Apart from AGFI and NFI which are approximately 0.9 (acceptable limit), CFI and TLI are above 0.9 thus indicating that the model passed the incremental fit. The parsimonious model fit result is acceptable. The overall model fit diagnosis shows that the model is good and acceptable. Table 2 reports results on the goodness of fit of the SEM model. The results show good fit for the entire three model's fit tests. The Chi-Square value is not significantly different from the degrees of freedom and their ratio passes the parsimonious model fit test (1.131).
With the exception of entrepreneurial self-efficacy, all the constructs show significant relation with entrepreneurial intention. The results show that locus of control has significant but negative relation with entrepreneurial intention Table 3 shows standardized and unstandardized coefficients for the constructs. The results show very significant relation between attitude towards entrepreneurship, subjective norm, locus of control and environmental support. Entrepreneurial self-efficacy does not significantly predict student entrepreneurial intention.
We measured five latent constructs on entrepreneurial intention using the questionnaire carefully developed after reviewing several literatures on the subject. The results show that attitude towards entrepreneurship is the construct that significantly contributes most (68%) in explaining variations in student entrepreneurial intention. This is followed by subjective norm (30%) and environmental support (19%). Interestingly, locus of control shows significant negative relationship with entrepreneurial intention. Overall, the model (all the five constructs) explains 89% of variations in student entrepreneurial intention and this is higher than similar and previous studies. This can be seen in the path model found in Figure 1. In Trivedi's study in 2016, the adjusted R for  the regression of ATB, SN, PBC and university environment and support on entrepreneurial intention was 0.69 which indicated that the model was highly significant since more than 69% of variation in entrepreneurial intention could be explained by the four predictors.

Discussion
This present study analyzed the entrepreneurial intention among the MBA students by reviewing literature on factors influencing Entrepreneurial Intention in line with the Theory of Planned Behaviour as the basic framework and two other variables; Locus of Control and Environmental Support. The results of the present study answer three (3) hypotheses and reject two (2) hypotheses.
The data for the study were obtained from MBA students from two of Ghana's private universities and the results are revealing.

Attitude towards entrepreneurship and entrepreneurial intention
From the results, it can be said that Attitude towards Entrepreneurship influences the entrepreneurial intention among the MBA students. These results mirror other studies by Armitage and Conner (2001) and Kim and Hunter (1993) whose studies have revealed a positive relationship between attitude towards entrepreneurship and entrepreneurial intention. Attitude Towards Entrepreneurship has statistically positive influence on entrepreneurial intentions among tertiary students (Buli & Yesuf, 2015;Ferreira et al., 2012). Trivedi (2016) also saw a strong and highly significant relationship between attitude and entrepreneurial intention. Surprisingly, Zhang et al. (2015) study which was conducted in the USA failed to generate a significant impact on entrepreneurial intention. In fact, we found a high score in the attitude toward entrepreneurship which incidentally happens to be the highest contribution to the entrepreneurial intentions in our model. Thus, this result showed that the influence of attitudes on intention has high explanatory power and extremely important for increasing entrepreneurial intention. Hence, we can argue that the MBA students are more independent and desire to be their own bosses in the near future with respect to their career path, which has the potential of curbing the problem of unemployment in the long run. Bosma and Kelley (2018) recognize that whenever employment opportunities and well-trodden career paths are scarce, creating a business is one of the few available avenues toward economic prosperity. They posit that the rate of people with the intention to start a business can exceed 60% in developing countries.

Locus of control and entrepreneurial intention
According to this study, LOC registered a significant but negative impact on the entrepreneurial intention among the MBA students. This finding sharply contradicts a study by Kristiansen and Indarti (2004) who reported a positive but insignificant relationship between LOC and entrepreneurial intentions among Indonesian students. The authors however, in the same study established a negative and insignificant relationship between LOC and EI among Norwegian students. According to Vodă and Nelu (2019), prior empirical research studying the relationship between LOC and EIs in European countries has produced contradictory results. For example, Rajh et al. (2016), in studying the entrepreneurial intentions of 1200 respondents from some European countries, found a positive but insignificant connection between LOC and entrepreneurial intention. Also, Figure 1. Popescu et al. (2016) in examining undergraduate and master students in Romania found a positive but insignificant relationship between LOC and entrepreneurial intention. Luthans et al. (2006) asserted that individuals with an internal locus of control are likely to positively face challenges and hindrances and they resolve those inhibitors by seeking productive solutions by displaying achievement motivation.

Entrepreneurial self-efficacy and entrepreneurial intention
In spite of the overwhelming empirical evidence to support a positive relationship between ESE and Entrepreneurial Intentions (Krueger et al., 2000;Luthje & Frank, 2003;Puni et al., 2018), this study revealed that Entrepreneurial Self-efficacy has no effect on the entrepreneurial intention among the MBA students. This is contrary to Sesen's (2012) study that entrepreneurial self-efficacy has a significant impact on entrepreneurial intentions. Laguna (2013) stated that self-efficacy is positively related to entrepreneurial intention. Also, Douglas and Fitzsimmons (2013) established a strong relationship between ESE and the intrapreneurial intentions of MBA students. However, some authors (e.g., Boukamcha, 2015;Kolvereid & Isaksen, 2006) found no evidence of such a relationship. In our study, 60% of the respondents were in the public sector as reported in the descriptive section of the analysis. In Ghana, job security in the public sector is normally guaranteed which inhibit the propensity to venture into entrepreneurial activities. According to Nowinski and Haddoub (2019) a positive attitude towards entrepreneurship is a necessary but not sufficient in fostering entrepreneurial intentions and they further emphasized that a positive attitude towards entrepreneurship needs to be supported by ESE and inspiring role models. According to McGee and Peterson (2019) people who believe in their ability to undertake certain activities are more likely to be successful in those activities.

Environmental support and entrepreneurial intention
It came out from the study that Environmental Support shows a significant relation with Entrepreneurial Intention. According to Sesen (2012) environmental forces (e.g., access to capital) has a significant impact on entrepreneurial intentions. According to Luthje and Frank (2003) if students realize a hostile environment for business founders due to perhaps the banks do not readily provide loans or because they perceive the state laws as being overly restrictive, they are less likely to venture into entrepreneurship. In Ghana particularly, the hostile and turbulent nature of the environment at times poses a challenge not only to nascent entrepreneurs but even to the existing ones. This is because entrepreneurial finance accessibility is a critical ingredient for success and one of the most important challenges facing entrepreneurial ventures is access to capital at realistically optimal interest rates. According to Jena (2020), the support entrepreneurs get from the environment (e.g., Mentor, Government and Financial institutions) could influence entrepreneurial intentions.

Subjective norm and entrepreneurial intention
The study revealed that Subjective Norm has positive relationship with EI, which contradicted the findings of other researchers (Autio et al., 2001;Krueger et al., 2000;Linan & Chen, 2009;Maes et al., 2014) which reported a non-significant relationship. Hiatt et al. (2009) established that social norms that inspire regulations can strongly affect organizational formation and failure. However, this study extends Ferreira et al. (2012)'s studies on the effect of Subjective Norm, who established that SN has a significant relationship. Interestingly, Nguyen et al. (2019) did not find the linkage between social norms and entrepreneurial intentions when they examined the factors affecting EIs among the youths in Vietnam. According to Moriano et al. (2012) subjective norms were significantly related to intentions in only two out of the six countries in their study.

Implications for theory and practice
This study was able to apply other operational measures than proposed by Ajzen (1991Ajzen ( , 2002 to test the robustness of the model in predicting entrepreneurial intentions as suggested by previous studies (Engle et al., 2010).
This study highlighted ATE as one of the important determinants of our framework; hence, entrepreneurial attitudes may be influenced by educators, policymakers and successful business owners.
As the adage goes, "every organization is as good as the people in the organisation", hence the selection and socialization of leaders (especially top management) and most importantly lecturers who share and endorse the idea that entrepreneurs make the difference in every society is very crucial for the continued relevance of private universities. We believe that by virtue of the superiorsubordinate power relationship in Ghana, top management in the academic institutions can inspire entrepreneurial intentions and behaviours among their subordinates (e.g., lecturer) which will eventually cascade down to the students.
Since almost all of our respondents fall within the category of the working class, there are practical implications for private, public and voluntary businesses and industries. It is important for management in the various organisations to instil the entrepreneurial spirit and proclivity in the workers to cause transformation and a turnaround in their respective institutions. This orientation is important not only for the incumbent employees but also for prospective ones in their recruitment and selection.
According to Abadi et al. (2021, p. 3), "people are influenced by norms as they go through and interact with those who are around them in social circumstances, the extent to which people face advocating or inhibiting norms determine the likelihood that they take an action or not". To the extent that SN has proved an important and significant determinant of entrepreneurial intention is refreshing. The positive effect of subjective norms on entrepreneurial intention is probably due to the prevalence of favourable reactions that the students give to the influences of important people, giving rise to positive intentions. We believe these influences have the potential of creating businesses in Ghana to curb the problem of unemployment in Ghana. According to Asiedu and Donkor (2018), respect for the views of elders and people in one's close circles is considered an important virtue in Ghanaian culture. These authors emphasize that it is a belief in Ghana that family leaders, community leaders and religious leaders are a repository of great knowledge and wisdom and their counsel is normally held in high esteem. Hence if family associates encourage the respondents to move into an entrepreneurial venture, they are most likely to concede to such an advice (pursuing a career as an entrepreneur). We can infer from the proposition of Asiedu and Donkor (2018) that societal norms can be institutionalized when they are accepted by individuals and groups and the motivation for the transition into entrepreneurial venture can be enhanced, all other thing being equal.
For policymakers, the findings indicate that high interest rate is counterproductive and serve as a disincentive in the encouragement of entrepreneurship. Most especially for these categories of respondents (MBA students) who are already in gainful employment and feel "safe" in their comfort zones, higher cost of capital can further push them away from an entrepreneurial dream. Probably, the government and other policymakers can collaborate with the private universities to institute start-up incentives in order to improve the students' entrepreneurial inclination. In recent times, private universities in Ghana have been lamenting over government's negligence.

Conclusion
The paper examined the entrepreneurial intentions among MBA students of two private universities in the Brong Ahafo region of Ghana. Specifically, we analyzed how locus of control, entrepreneurial self-efficacy, environmental support, subjective norm and attitude determine students' entrepreneurial intention.
We confirm that TPB-based variables can be adopted in an area in Ghana's educational sector which seems to be "forgotten" with respect to entrepreneurial intentions research; thus contributing and deepening to previous TPB-based research on EI.
This study registered statistically significant explanatory power of 89% of the variations in Entrepreneurial Intention, due to ATE, SN, ESE/PBC, ES and LOC which is considered to be more robust given most research fall short of this number. Our study adds to existing literature in an area which has barely produced literature in universities in Ghana (particularly private universities) and the overreliance on the developed countries, through empirically testing of Azjen's TPB-based variables in the context of private universities in Ghana.
This study revealed that the Entrepreneurial Intentions of MBA students are influenced by ATE. This implies that pragmatic strategies and tools should be incorporated in the private MBA curriculum to promote student's attitudes towards job creation. Unfortunately, the total number of entrepreneurship courses taught in each of the two institutions surveyed is not more than two subjects. According to Bogatyreva et al. (2019), on the average, students with entrepreneurial intention during schooling are approximately three times more likely to start a business after school, as compared to students without intention. Besides, Entrepreneurial Intentions are the most proximal predictor of individual academics' engagement in entrepreneurship. However, the universities in Ghana offer few courses related to entrepreneurship. The findings can be used to guide universities in Ghana, government and other stakeholders on how to stimulate entrepreneurial intentions among students.
Entrepreneurial self-efficacy is one of the most important determinants of entrepreneurial intention but this study revealed non-significant relationship; hence, it is recommended that policymakers should allocate resources in a manner that promotes the entrepreneurial selfefficacy of MBA students in the private universities. According to Phong, Thao, and Nguyen (2020, p. 19), "intervention programmes like critical thinking, negotiation, presentation, time management, networking, cross-cultural awareness skills, or even down-to-earth activities related to business skills like basic golf, professional dining, and grooming" can propel entrepreneurial self-efficacy.

Limitations and future research
Our data were based purely on quantitative and cross-sectional data. For instance, the crosssectional nature of this study may not give room for stronger causality inferences. We, therefore, suggest that future research should apply both quantitative and qualitative research approaches, and also, the application of longitudinal research designs, as it can make significant contributions.
We examined the MBA students' perceptions in relation to their future entrepreneurial intentions, but not actual behaviours. Since intentions may not necessarily lead to actions, further longitudinal studies about the factors associated with entrepreneurial intentions and to establish whether the respondents actually "walk the talk". In fact, entrepreneurial intentions, however, are only a first step toward entrepreneurial action and eventual business ownership. According to Krueger et al. (2000), intention-based models examine the intent, but not the timing of venture creation. Interestingly, it may take time after intent metamorphoses before a new venture opportunity is even recognized.
The respondents in our study were primarily students from only two private universities in Ghana, which may render our findings less generalizable to other higher institutions of learning. However, our framework can be adopted and applied in different context for future studies in order to verify the authenticity of the model.
Entrepreneurial education undoubtedly is one of the critical to the success in the development of entrepreneurial competences; hence, the need for policymakers to integrate entrepreneurship education in the MBA programs of private universities, since a number of these courses have the potential of enhancing entrepreneurial intention and propensity. According to Puni et al. (2018), due to the rising rates of unemployment in Sub-Saharan Africa and its attendant economic and social problems, stakeholders are embracing the concept of entrepreneurship education as a major conduit in shaping the quality of human capital for full employment.
Lastly, the constructs used in this study are not the only variables in the determinants of entrepreneurial intentions. Future studies might integrate broader constructs to measure entrepreneurial intentions among university students.