The effect of perceived barriers on social entrepreneurship intention in Malaysian universities: The moderating role of education

Article history: Received: November 26, 2017 Received in revised format: March 31, 2018 Accepted: April 4, 2018


Introduction
It has been highlighted that social, economic and environmental aspects of development would pose the greatest challenges in the 21st (Jørgensen et al., 2009). Therefor governments are faced with many problems that they are not well equipped to deal with such rising social needs and issues (Lumpkin et al., 2013). Also, it is believed that the social entrepreneurship represents a new structure and ideology that can capture social, economic and environmental values for society's development (Nga & Shamuganathan, 2010;Noruzi et al., 2010;Zebrowski, 2009;Davari & Farokhmanesh, 2017). Further-more, it is believed that social entrepreneurship provides solutions for social, economic and environmental problems through actors, individuals, and organizations acting as new pioneers (Santos, 2012;Tan & Yoo, 2015;Verreynne et al., 2013). In this regard, the emergence of social entrepreneurs is seen as a social change agent who is creating social value (Simha & Carey, 2012;Swanson & Di Zhang, 2010). Intention is an integral part of any social entrepreneur's life, as he/she seeks to eliminate a social need, a social problem, and goes far beyond his/her purely commercial entrepreneurial spirit (Austin, 2006;Zahra et al., 2009). Even Mair and Marti (2006) speak about an intention formation stage, which is to some extent in line with Ajzen's (1991) theory. Students also need to have such feelings and intentions to initiate social businesses and to continue this behavior (Salamzadeh et al., 2013). Additionally, there is an increase in the creation of awareness about the importance of entrepreneurship education in academic and co-curricular activities.
In Malaysia, social entrepreneurship is in early stages and requires a proper policy to promote social entrepreneurship and its education (İlhan Ertuna & Gurel, 2011;Kurniawan et al., 2012;Terjesen et al., 2011). Researchers have pointed out that majority of the Malaysians are not aware of the concept of social entrepreneurship and how it can benefit them, thus there is a need to understand this phenomenon in Malaysian context (Al-Mamun et al., 2012;Sarif et al., 2013;Terjesen et al., 2011). However, some educational institutions in Malaysia are involved in the promotion of social enterprises at different levels. but, some academic institution and consultancy systems in Malaysia have not yet undertaken a great deal of research and capacity development in social enterprises, hence, there is an enough room for the improvement in this area (MaGIC, 2015). To promote entrepreneurship we need to know the barriers that affect entrepreneurship intention to overcome the barriers (Yaghoubi, 2010). According to Ajzen's theory Perceived barriers are one of the factors that can affect on social entrepreneurship intention among students (Kirby & Ibrahim, 2011;Salamzadeh et al., 2013).
Few studies have examined the differences in entrepreneurial intentions and perceived barriers to business creation among students (Giacomin et al., 2011). To accomplish this, we believe it is important to study students' perceived barriers on social entrepreneurial intention. The theme "perceived barriers" describes how young staff's self-efficacy could be negatively influenced by social exclusion, forecasted on poor academic achievement or debilitating psychological states, which may influence on their capabilities to succeed in enterprising activities (Lucas & Cooper, 2005;Mclellan et al., 2009). Furthermore, limited knowledge is available on the perceived barriers that reduce the social entrepreneurial intentions of the students. These barriers can be the lack of support, lack of adequate knowledge and competency, lack of self-confidence and lack of resources (Shinnar et al., 2012). Studies have already acknowledged the negative effect of these perceived barriers on social entrepreneurial intentions (Giacomin et al., 2011;Shinnar et al., 2012), but still, the role of social entrepreneurial education can play in the mitigating the effect of these barriers is unclear.
This study has used the theory of planned behavior as a theoretical framework to investigate the effect of these barriers on the social entrepreneurial intention and the moderating role of social entrepreneurial education to mitigate the effect of perceived barriers on the social entrepreneurial intentions.

Social entrepreneurship intention
In the entrepreneurship literature, prior studies demonstrated the role of intentionality as the best predictor of planned behavior (Ajzen, 1991), recognizing that starting a business is an intentional act and entrepreneurship is a planned behavior (Krueger et al., 2000). Intentions reflect the motivational factors that influence behavior and are a reliable indicator of how hard a person is willing to try and how much effort he/she makes to perform a behavior (Ajzen, 1991;Austin, 2006). The intent is not only connected with the desire to work as an entrepreneur, but also their aspiration towards entrepreneurship (Rantanen & Toikko, 2013). Students also need to have such feelings and intentions to initiate social businesses and to continue this behavior (Salamzadeh et al., 2013). The available literature on the entrepreneurial intentions has shown the interest of the researchers to use the theory of planned behavior to investigate the entrepreneurial intentions (Fretschner & Weber, 2013;Mair & Marti, 2006;Rantanen & Toikko, 2013;Salamzadeh et al., 2013;Tan & Yoo, 2015). This provides enough justifications to use the theory of planned behavior in the current study to investigate the social entrepreneurship intentions. Social entrepreneurship intention refers to the intention of starting a social enterprise. The literature indicates that there is a need to adapt the existing educational system if social entrepreneurs are to be encouraged (Salamzadeh et al., 2013;Yunus, 2007).

Social Entrepreneurial education
With the increased social problems the world is facing, there has been a demand of increased social entrepreneurship related program (Brock, 2008;Miller, 2012;Nandan & Scott, 2013). These programs and activities are not only focusing on creating awareness and providing services for development but are also concerned with developing social entrepreneurial culture and skills (Abdou et al., 2010). In addition, for a fast developing country like Malaysia, there is an essential need of understanding the social entrepreneurship phenomenon (Sarif et al., 2013) especially when Malaysia is looking to move towards knowledge based economy (Xavier et al., 2010). Students are very crucial for developing cogent global sustainability practices and principles giving a balanced emphasis to the economic, environmental and social concerns a majority of these students will immediately contribute more to being a process of change in the delivery of public goods and social/environmental services. And they will have an important role in an economy, in particular for wealth creation (Thompson et al., 2000).
Despite the criticism, studies have shown that education influences students' inclination and behavior towards entrepreneurship (Gorman et al., 1997;Kolvereid & Moen, 1997;Min et al., 2012). Furthermore, Tracey and Phillips (2007) highlighted that there is an increase in a number of social entrepreneurs who are enrolling in universities to learn the skills for building sustainable ventures. In this regard, It has been highlighted that universities by developing social entrepreneurship education have the opportunity of not only acting as change agents but also are in a position to develop human capital that could be used for transforming the world and addressing the social challenges (Brock & Kim, 2011). Moreover, some articles emphasize the importance of collaborating with many partners in society when students learn to think and behave as social entrepreneurs (Jensen, 2014;Prieto, 2011Prieto, ). (1997 have indicated that there exist a causal relationship between education and (social) entrepreneurial behavior. Studies have suggested that higher education would influence student's intention of becoming entrepreneur (Brock, 2008;İlhan Ertuna & Gurel, 2011;Mair & Marti, 2006;Shankar, 2014) and local development and poverty reduction and the economic growth after they graduate, not as salary workers but as social entrepreneurs (O'Connor, 2013). While much research has focused on entrepreneurial intentions (BarNir et al., 2011;Ismail et al., 2009;Kuckertz & Wagner, 2010;Shinnar et al., 2014;Souitaris et al., 2007), a large amount of literature concerning the social entrepreneurship education has been produced (Brock & Steiner, 2009;Pache & Chowdhury, 2012;Salamzadeh et al., 2013) and some research studied the effect of Social entrepreneurship education on intention (İlhan Ertuna & Gurel, 2011;Kirby & Ibrahim, 2011;Salamzadeh et al., 2013) but there is little study to investigate the direct effect of perceived barriers on social entrepreneurial intention, by the moderating role of Education.

Perceived Barriers
Perceived control of behavior is one of the dimensions of the theory of planned behavior related to the perception of an individual regarding the easiness or difficulty of a certain behavior. In fact, it proposes that a person's intention to perform the behavior will increase with his/her perceived behavioral control. This suggests that people will try to start a business if they believe they have the ability to do so and the activity is deemed to be socially acceptable (Kirby & Ibrahim, 2011).
Some people believe that some barriers (ex: fear of failure, irregular income, and having to work too many hours) have negative effects on student intention and these barriers should be taken into consideration when developing entrepreneurship education programs.
In this regard, previous studies conducted in the Malaysian context revealed that students lack explicit knowledge of entrepreneurship and social entrepreneurship and the intention to start their own entrepreneurial venture (MaGIC, 2015;Terjesen et al., 2011;Zakaria, 2011). Furthermore, a study referred to some problems relates social entrepreneurship programs among the Malaysian students. These problems do not have motivating students towards social entrepreneurship programs, the problem of different perceptions among the students, lack of knowledge and skills of students have also slowed down the process of social enterprise creation, lack of active involvements of government, private agencies, local and national companies in social enterprise process and lack of specialized training for the educators. Further, despite the implementation of various policies and programs by the Malaysian government to enhance entrepreneurial activity in the country, students are still lacking the awareness, confidence, abilities, and competencies required for the success of (social) entrepreneurial venture (Kurniawan et al., 2012;Terjesen et al., 2011).
Barriers to entrepreneurship can include difficulties in obtaining institutional support for aspiring social entrepreneurs, receiving family support, securing financing from lenders, building a relationship with suppliers, and/or a solid customer base. Lüthje and Franke (2003) indicate that the perceived availability of support such as access to "qualified consultants and service support for new companies" (p. 147) have positive impacts on entrepreneurial intentions. This study would expect that the perceived absence of such support may act as a barrier. Miller (2012) proposed that resource providers entrepreneurs depend on (i.e., consultants, lenders, suppliers, customers) frequently make decisions under uncertainty without access to complete information. They may, therefore, be especially vulnerable to the influence of challenges social entrepreneurs. Specifically, Onwuegbuzie and Collins (2007) found the fear of failure to be negatively related to entrepreneurial propensity. Furthermore, Miller (2012) identified the lack of knowledge and competency is another potential barrier for the social entrepreneur intentions. Additionally, Tukamushaba et al. (2011) argued that the lack of self-confidence and resources are also the potential barriers to the social entrepreneurial intentions. Moreover, According to Tukamushaba et al. (2011) the main barriers, as perceived by students who had yet to start their own venture, would appear to be the lack of awareness or information/knowledge. Of those who claimed that they would be interested in setting up a social enterprise, most indicated that they would need support/encouragement, particularly training and mentoring. However, the effect of social entrepreneurial education on social entrepreneurship intention to mitigate perceived barriers is still unclear. Thus, on the basis of above discussion study concluded following hypothesis: H1: lack of support has a negative effect on social entrepreneurial intention. H2: lack of knowledge has a negative effect on social entrepreneurial intention. H3: lack of competency has a negative effect on social entrepreneurial intention. H4: lack of self-confidence has a negative effect on social entrepreneurial intention. H5: lack of resources has a negative effect on social entrepreneurial intention. H6: Social entrepreneurial education moderates the relationship between the perceived barriers and social entrepreneurial intention.

Sampling strategy
We used multilevel sampling design with purposive sampling scheme as highlighted by Yunus (2007) that multilevel sampling involves 'sampling in all levels being purposive or sampling in all levels being random'. Purposive sampling technique is preferred for the selection of the universities. Criteria for the selection of the universities were the research universities. Research universities have more appropriate resources for encouraging students to be social entrepreneurial through the spinoffs and commercialization of their research opportunities. This technique is preferred as prior researchers have also used the purposive sampling technique in collecting data (Jensen, 2014;Tukamushaba et al., 2011;Yasir et al., 2013). For the analysis in the current study, students of the research universities of Malaysia were used. We collected data from 288 students from Malaysian research universities. Following sections highlights the demographic analysis using frequency tests of the respondents. Majority of the respondents belongs to age group of 18-26 years (55.6%). This is followed by respondents belonging to the age category of 26-30 years (35.2%). The rest belongs to the other age groups. Gender wise analysis indicated that majority of the respondents were female (58.3%). The education level indicates that majority of the respondents were under graduates (47.8%). Second highest frequency was the respondents having master's level education (22.5%).

Measures
Perceived barriers are independent variable in this study. Perceived barriers refer to the degree to which students of universities consider the difficulties in starting social entrepreneurship. We used five dimensions of the independent variable of perceived barriers. These dimensions are lack of support, lack of knowledge, lack of competencies, lack of self-confidence and lack of resources. We operationalized each dimension through 3 items on five points likert scale where 1 represents strongly disagree to 5 strongly agree. The scale was adapted from Giacomin et al. (2011).
We  Salamzadeh et al., 2013). Researchers have also highlighted the linkage between social entrepreneurial traits and intentions of students to start an enterprise (Bridges, 2008;İrengün & Arıkboğa, 2015;Mair & Noboa, 2003;Nga & Shamuganathan, 2010;O'Connor, Burnett, & Hancock, 2009;Prieto, 2011) with education acting as a moderator (İlhan Ertuna & Gurel, 2011). For the present study, social entrepreneurship education is taken as a moderator to examine the intentions of students to start social enterprise through planned behavior theory. Social entrepreneurship education is coded dichotomously where "0" indicated participation in social entrepreneurship course and "1" refers to student who does not participate in social entrepreneurship course.

Measurement Model
As a required step in testing the conceptual models, the suitability of the computed variables must be assessed. Confirmatory factor analysis was conducted by using AMOS 18. The results of the CFA indicated that all of the values are within the acceptable ranges as shown Fig. 1 for eliminating the items were set on the basis of the factor loadings and the residual values of the each item. The factor loadings >.50 was selected to retain the items. Fig.1. showed

Fig.1. Measurement Model
Convergent validity is the construct indicators that reflect a large amount of the mutual proportion of variance among factors. It determines the amount of correlation among the measures of the same concept. Convergent validity deals with construct loadings, average variance extracted (AVE) and construct reliabilities. Average variance extracted is the sum of square of standardized factor loadings to represent how much variation in each item is explained by latent. The average variance extracted is the average percentage of variation explained by the measurement items in a construct. The standard value of AVE is .50 or greater. Table 1 shows the average variance extraction of each construct and results showed that all the constructs have more than .50 of average variance extraction, that shows all the constructs have sufficient amount of convergent validity. The range of Average Variation Extractions is 0.549-0.650. The threshold value of the construct reliability is .70 or above. Table 1 shows that all the constructs have the adequate reliability of all constructs ranges from .784 to .920. Therefore, the current study does not violate the convergent validity of the constructs.
Discriminant validity is referred to the extent to which an instrument contains a construct that was truly distinct from all others. Discriminant validity is the degree to which similar constructs have distinct values. In this type of validity, the responses are measured without cross loading in terms of latent constructs. Discriminant validity is violated when the correlation among exogenous constructs is more than 0.85. In discriminant validity, the value of the square root of average variance extraction should exceed than the value of inter-construct correlations. Table 1 shows the inter-construct correlations.
Results indicate that all the constructs have adequate discriminant validity as the square root of average variance extracted is greater than the inter-construct correlation of each variable and also the values of inter construct are less than .85. It means the results provide sufficient evidence of the discriminant validity of the constructs.

Structural Model
The proposed structural model consists of 5 hypotheses, as indicated in Fig. 1. Our first hypothesis states that H1: lack of support has a negative effect on social entrepreneurial intention. So the standardized path coefficient of lack of support and social entrepreneurial intention was 0.01 with p value above than 0.05. Thus hypothesis 1 was not supported and the study does not establish a relationship between the lack of support and social entrepreneurial intention.
Our second hypothesis states that H2: lack of knowledge has a negative effect on social entrepreneurial intention. So the standardized path coefficient of lack of knowledge and social entrepreneurial intention was 0.00 with p value above than 0.05. Thus hypothesis 2 was not supported and study does not establish relationship between the lack of knowledge and social entrepreneurial intention.
Our third hypothesis states that H3: lack of competency has negative effect on social entrepreneurial intention. The standardized path coefficient of lack of competency and social entrepreneurial intention was -0.11 with p value less than 0.05. Thus hypothesis 3 was supported and study establishes negative relationship between the lack of competency and social entrepreneurial intention.
Our fourth hypothesis states that H4: lack of self-confidence has negative effect on social entrepreneurial intention. The standardized path coefficient of lack of self-confidence and social entrepreneurial intention was -0.40 with p value less than 0.05. Thus hypothesis 4 was supported and study establishes negative relationship between the lack of self-confidence and social entrepreneurial intention.
Our fifth hypothesis states that H5: lack of resources has negative effect on social entrepreneurial intention. The standardized path coefficient of lack of resources and social entrepreneurial intention was -0.42 with p value less than 0.05. Thus hypothesis 5 was supported and study establishes negative relationship between the lack of resources and social entrepreneurial intention. Fig. 2 shows the direct relationship We used group analysis to test the moderation effect of social entrepreneurial education between the relationship of perceived barriers and social entrepreneurial intentions. For this purpose, study runs two models on the two different respondents groups. First group was the students who have already taken the social entrepreneurial course during their studies and the second group included those students who did not take social entrepreneurial course during their studies. The difference in the Chi-Square of both models tells the moderation effect of the social entrepreneurial education. Results of the group analysis were mentioned in the Table 2.  Table 2 shows the difference in results between both groups. Hypothesis 1 was not supported by two groups. It means that lack of support has a negative effect on social entrepreneurial intention was not supported by a student who educated and non-educated on social entrepreneurship. Hypothesis 2 was not supported by the students who took the social entrepreneurial course during their studies. This means the students who have passed the social entrepreneurial course do not perceive lack of knowledge as a barrier for entering in social entrepreneurial activities in future. However, the lack of knowledge affect negatively to the social entrepreneurial intentions was supported by the students who did not take the social entrepreneurial course during their studies. Hypothesis 3 was supported by two groups. It means that lack of competency has a negative effect on social entrepreneurial intention by educated students and non-educated ones on social entrepreneurship. Hypothesis 4 was supported by two groups. It means that lack of self-confidence has negative effect on social entrepreneurial intention by educated students and non-educated ones on social entrepreneurship. Hypothesis 5 was supported by two groups. It means that lack of resources has negative effect on social entrepreneurial intention by educated students and non-educated ones on social entrepreneurship. Chi-square difference test was applied to test the significance of the moderation effect. Table 3 shows the results of the Chi-square difference test for both models. Chi-square value for the model on the data of the students who took the social entrepreneurial course during their studies was 891, with 241 degrees of freedom and for the non-educated group was 561 with degrees of freedom 199. The difference between the Chi-square was 330 with 42 degrees of freedom. T value for the test was 7.86 (must be above 1.96) with the p value of 0.000 (regression: must be less than 0.05). This indicates the existence of moderation effect of social entrepreneurship education between the relationship of the perceived barriers and social entrepreneurial intentions.

Discussion and conclusion
The aim of this study was to investigate the relationship between the five perceived barriers with the social entrepreneurial intention of the students in Malaysian Research University along with moderation effect of social entrepreneurial education. We have considered the five types of the perceived barriers; namely lack of support, lack of knowledge, lack of competency, lack of self-confidence and lack of resources. The results indicated that overall students considered the lack of competency, lack of self-confidence and lack of resources as the barriers influencing on the social entrepreneurial intentions of the research university students in Malaysia. However, the lack support and lack of knowledge were not considered as the potential barriers for the social entrepreneurial intentions of the students. Results also indicated that the social entrepreneurial education moderates the relationship between the perceived barriers and social entrepreneurial intentions of the students. The students who passed the social entrepreneurial course during their graduation are less likely to vulnerable to the perceived barriers to enter social entrepreneurial activities in future as compared to those who did not attend social entrepreneurship programs as a course during their study period. This implies that teaching of social entrepreneurial course at the university level can help to reduce perceived barriers that may reduce the entrepreneurial intentions of the university students. Thus, the Malaysian government in order to enhance the social entrepreneurial activities in the country could strategize the teaching of social entrepreneurial course as a compulsory subject at university level to reduce the perceived barriers among university students in Malaysia. This study recommends that a cross country study among developing countries and developed and developing countries to realize barriers of social entrepreneurship intention in universities.