Influence of perceptual and demographic factors on the likelihood of becoming social entrepreneurs in Saudi Arabia, Bahrain, and United Arab Emirates – an empirical analysis

Abstract

A growing body of literature contributes to empirical examination of social entrepreneurship (S-ENT) intention formation. However, the extant literature overlooks the influence of perceptual factors in explaining the individual’s propensity to engage in S-ENT. Drawing on this gap and taking as starting point the literature on cognitive entrepreneurship, the study aims to identify and examine the influence of individual perceptual factors on the likelihood of becoming a social entrepreneur. A questionnaire survey was conducted to collect the data on adult population from three Gulf Cooperation Council (GCC) countries. The results of binomial logistic regression revealed the role of five individual perceptions (confidence in one’s skills and knowledge, ability to perceive opportunities, knowing other social entrepreneurs, fear of failure, and perceived social passion) as influencing the likelihood of becoming a social entrepreneur. Additionally, the economic and demographic variables were also analyzed to reveal the significant differences. While addressing the shortcomings of previous studies, the results provided the groundwork for furthering research on S-ENT intentions and behavior in any similar context, thus providing the factors that need to be focused for enhancing the culture of S-ENT.

Keywords:

• social entrepreneurship
• social entrepreneurs
• perceptual variables
• social passion
• entrepreneurship behavior

1. Introduction

Social entrepreneurs are emerging as distinctive entrepreneurs aiming to create social value (Teck et al., 2020). Despite being characterized by unique challenges like double or triple bottom line objectives, primacy of social goals, non-distribution constraints, barriers to market entry, the uncertainty of returns, etc. (Kuckertz & Wagner, 2010; Satar & John, 2019). The practice of social entrepreneurship (S-ENT) is growing across the globe (Dwivedi & Weerawardena, 2018). Meanwhile, the scholarly methodologies in S-ENT are gradually shifting away from case or anecdotal studies to more rigorous qualitative and quantitative approaches to turn an essentially practitioner-led S-ENT endeavor into an objective field (Bosma et al., 2016; Lall & Park, 2022; Lepoutre et al., 2013). In fact, the goals of many S-ENT approaches till date can be defined as the need to understand where, why, and how social entrepreneurs identify and exploit social value creation opportunities (Calderón-Milán et al., 2020; Kruse et al., 2021). In this regard, the increasing focus on cognitive approaches has been shown to be beneficial in explaining social entrepreneurial behavior at both the individual and collective levels (Krueger et al., 2011; Linan et al., 2011). Perceptions are a cognitive construct that has been shown to predict or explain entrepreneurship behavior, such as the likelihood of starting a new venture (Arenius & Minniti, 2005; Linan et al., 2011). According to the literature, perceptual factors play a significant role in a person’s propensity to engage in social entrepreneurial activity (Barba-Sánchez et al., 2021). For instance, it has been discovered that the perceptions of entrepreneurial self-efficacy, the ability to spot opportunities, knowing other entrepreneurs, and the fear of failure are strongly associated with the chances of starting a new business (Arenius & Minniti, 2005). Further, in order to identify the underlying reasons for differences in perception, researchers have focused on the socio-demographic factors (e.g., gender, age, educational level, employment status) and other contextual factors as well (Camelo-Ordaz et al., 2016; Langowitz & Minniti, 2007).

S-ENT is a fluid and rapidly evolving concept throughout the world especially in the developing countries and countries with relatively greater income differences like GCC. Currently, there are those that define the social enterprise as a purely commercial entity, which is profit driven and solely accountable to its shareholders. However, non-profit actors who are almost exclusively concerned with addressing a particular social ill are accountable only to their donors. Thus, identification of the factors that influence the individuals to become social entrepreneurs in the gulf region becomes important, and hardly discussed before.

Additionally, perceptions play a role in mediating the association between entrepreneurial intention and traditional characteristics including age, household income, employment status, and education levels (Camelo-Ordaz et al., 2016). Although the literature on cognitive entrepreneurship has concentrated on the impact of specific perceptions on start-up intentions, yet social entrepreneurs are thought to be less influenced by typical opportunity identification and exploitation characteristics (Short et al., 2009) possibly due to influence of their unique entrepreneurial orientation over entrepreneurial intentions (Kraus et al., 2017; Naveed et al., 2021; Satar & Natasha, 2019). This calls for research for:

1. whether the established perceptions of entrepreneurial self-efficacy, the ability for opportunity identification, knowing other entrepreneurs, and the fear of failure have an equivalent impact on individual’s propensity to engage in S-ENT context and

2. if there are any individual perceptual factor(s) unique to the S-ENT context?

Although a growing body of literature contributes to empirical examination of social entrepreneurial intention (SEI) formation (Hockerts, 2015; Pham et al., 2022). however, the literature on SEI is more or less fragmented, yielding a variety of outcomes and paradoxes (Lall & Park, 2022; Short et al., 2009). Meanwhile, with the triple bottom-line objectives (Satar, 2022), the S-ENT venturing is predictably a socially embedded process (Anderson & Jack, 2002; Satar & John, 2019). As a result, subjective perceptions of one’s environment and one’s relative position in that environment are critical considerations in S-ENT venturing (Lepoutre et al., 2013; Satar, 2020). SEI literature thus overlooks the importance of perceptual factors in shaping the individual’s propensity to venture in S-ENT context. Therefore, the current study attempts to fill this gap by drawing on the above-described cognitive typology. Hence, this study specifically relates the cognitive method to examination of S-ENT antecedents by examining the individual demographic and economic aspects, along with perceptual variables. By gathering and analyzing the general adult population data across three countries, the current study provides the opportunity for the aggregate analysis of S-ENT individual perceptions and creates the groundwork for subsequent analyses of S-ENT intentions and behaviors. Additionally, the study goes a step further in examining the influence of socio-demographic indicators like gender, income, and education levels, etc., which can therefore, enrich the literature on theory and policy building in S-ENT context.

1.1. Theoretical background

Most of the studies over entrepreneurial intentions or social entrepreneurial intentions are based on Theory of Planned behavior, or entrepreneurial event model. However, both the theories mainly focused on personality characteristics, as well as the events in the life of the individuals; however, the perceptual measures have hardly been catered by any researcher in the past. The study derived from the perspective that the research in psychology and experimental economics indicate that instead of thinking and weighing all things and options objectively, people rely on learning from behavior and attitudes of others, or from the phenomenon that captures their attention (Bandura & Walters, 1977; Fish, 2021). The psychological interpretations of the outer world that individual’s senses and minds gather are known as perceptions (Foxall, 2020), which have hardly been addressed in the past by the researchers especially in the context of social entrepreneurial intentions. Existing and aspiring entrepreneurs’ sense, the effects of the external world through their motives and perceptions, resulting in attitudes and intentions that drive their behaviors (Bandura & Walters, 1977; Fish, 2021). Research in various contexts has emphasized the significance of cognitive processes in entrepreneurial activities from the standpoint of entrepreneurial behavior (Krueger et al., 2011; Linan et al., 2011). However, S-ENT is a socially embedded phenomenon, one’s subjective judgments of his or her surroundings and their place within it are crucial factors in his/her likelihood of becoming a social entrepreneur (Dwivedi & Weerawardena, 2018). As a result, the current study fulfills the gap in the body of knowledge by investigating, on a broad scale, the role of perceptions as cognitive variables determining the individual’s propensity to become a social entrepreneur, which has not been studies before especially in the context of the Arab world.

1.2. Cognitive approach to entrepreneurial intentions and behaviours

The evolution of entrepreneurship literature can be analysed from two perspectives; the entrepreneur’s traits or personal characteristics and the impact of economic, social, cultural, and political contextual elements (Aldrich & Wiedenmayer, 1993). However, these approaches have been criticized from a theoretical standpoint due to their conceptual and methodological flaws, as well as their lack of explanatory power primarily because they neglect behavior as a result of person–situation interactions (Krueger et al., 2011). As a result, the cognitive approach has become increasingly important to the study of the social entrepreneurship behavior (Mitchell et al., 2004). The evolution of S-ENT research could be roughly tracked along comparable patterns. For example, the S-ENT literature has evolved as “individualism,” with a clear emphasis on social entrepreneurs as “heroic” individuals (Austin et al., 2006). The approach mainly focuses on identifying and distinguishing the social entrepreneurs based on their distinct traits like diverse levels of knowledge, non-profit motivation, social entrepreneurial competencies, social passion, change agents, and career aspirations (for example, see İ̇rengün & Arıkboğa, 2015; Zuhaib et al., 2022).

Furthermore, the scholars are attracted to find the relationships between personality traits and demographics with some social entrepreneurial behavior, such as Big Five Personality Traits, social responsibility, social legitimization of the social mission, entrepreneurship resilience, moral obligation etc. (Ernst, 2018; Nga & Shamuganathan, 2010; Sana et al., 2021). Meanwhile, there is huge diversity demonstrated by the social enterprises in their structure, strategies, processes, and missions (Satar et al., 2016). Accordingly, the importance of social contexts, socioeconomic, and demographic settings in the S-ENT process is now being examined in S-ENT literature (Nga & Shamuganathan, 2010; Satar, 2020).

Meanwhile, the emerging studies overwhelmingly focused on explaining the SEI formation with the analysis restricted to small samples generally made up of students (Ernst, 2018). The student samples possess serious limitations in predicting entrepreneurship intentions mainly in terms of a significant bias in the results (McGee et al., 2009). Secondly, SEI studies are predominantly grounded by theory of planned behavior and extending the Mair and Noboa (2006) intention model (Hockerts, 2017; Ernst, 2018). While these approaches have provided a kick-start to the exploration of certain cognitive and some promising intention constructs like empathy, moral obligation, pro-social orientation, entrepreneurship passion, perceived social support and self-efficacy (Chandra et al., 2021; Hockerts, 2015; Mair & Noboa, 2006). Thus, the SEI literature is more or less fragmented, yielding a variety of outcomes and paradoxes (see Baierl et al., 2014; Kruse et al., 2021; Naveed et al., 2021; Sana et al., 2021; Short et al., 2009). For example, the scholars in some contexts were unable to detect a link between empathy and S-ENT intention (Ernst, 2018). Even the trait approach did not prove useful in predicting the S-ENT intentions (Nga & Shamuganathan, 2010). Meanwhile, the traditional antecedents of entrepreneurial intention, such as opportunity identification and exploitation, have vastly different effects on social entrepreneurs (Short et al., 2009). Drawing on above arguments, below is a review of the literature devoted to analyzing the role of perceptual factors on the individual’s likelihood of becoming a social entrepreneur, from which we derive the corresponding research hypotheses.

1.3. Ability to perceive opportunities

One of the fundamental and distinguishing characteristics of entrepreneurial behavior is the perception towards opportunities (Langowitz & Minniti, 2007). There is strong empirical evidence that the ability to recognize opportunities is associated with a higher likelihood of becoming an entrepreneur (Arenius & Minniti, 2005; Langowitz & Minniti, 2007). A growing body of research indicates that social entrepreneurs are skilled at spotting opportunities. For instance, it is claimed that social entrepreneurs see opportunities where others perceive vacant buildings, unemployed people, and unused resources (Catford & Cited in Johnson, Sherryl, 2000). They produce fresh solutions to social issues. As an illustration, self-efficacy serves as a beneficial mediator in the relationship between entrepreneurial alertness and social entrepreneurs’ intention to start a business (Urban, 2020). Opportunity identification is a crucial task for all entrepreneurs, even though there are subtle distinctions between commercial and social business opportunities (e.g., Baierl et al., 2014). According to the patchy evidence, people who believe there are promising prospects to launch a social enterprise in the near future are more likely to do so than those who are unaware of/dismissive of such opportunities. As a result, the following hypothesis can be proven:

H1:

An individual’s ability to recognize social entrepreneurial opportunities has a positive influence on his/her propensity to become a social entrepreneur.

1.4. Knowing other social entrepreneurs (role model perception)

Entrepreneurship scholars believe that the role models can empower individuals to learn specific skills, gain the necessary knowledge (Terjesen et al., 2012), and thus, can trigger entrepreneurial behaviour (Minniti, 2005). In S-ENT context, although the role model influences on determining the SEI are expectedly more intense (see Nga & Shamuganathan, 2010; Sana et al., 2021), however, the examination and applicability of social learning theory (Bandura & Walters, 1977), possibly considering the role model influences, goes largely missing in S-ENT literature. Nevertheless, social entrepreneurs have emerged as heroic individuals (Austin et al., 2006), embedded in deep social networks (Satar, 2019), and demonstrating varied sources of motivation (Ganguli et al., 2021; Nga & Shamuganathan, 2010; Sana et al., 2021). With their stories and psychological circumstances, social entrepreneurs have become role models for an alternative professional career path in developing sustainable socioeconomic projects (Cohen & Katz, 2016). In effect, social capital, trust-building, and relationship ties are regarded as core characteristics of social entrepreneurs (Satar, 2019; Satar & John, 2019). Therefore, the following can be hypothesized:

H2:

People who know other social entrepreneurs are more likely to become social entrepreneurs.

1.5. Self-efficacy (confidence in one’s own skills and knowledge)

The concept of self-efficacy was investigated by developing entrepreneurial intention models based on the theory of “entrepreneurial events model” (Shapero & Sokol, 1982). As a result, perceived feasibility (also referred to as self-efficacy) is described as confidence in one’s own ability to carry out an entrepreneurial endeavour. Similarly, Ajzen’s (1991) “‘theory of planned behaviour’” was employed to describe how perceived subjective norms and perceived behavioural control, or self-efficacy influence the intention towards social start-up (Krueger et al., 2011). As a result, the studies have established that confidence in one’s skills and knowledge is a major motivator for deciding to start a new venture (Anjum et al., 2021; Terjesen et al., 2012).

While the studies investigating the S-ENT intention are emerging (Hockerts, 2015), the early indication of perceived feasibility and desirability as influencing factor for S-ENT intention was offered by Mair and Noboa’s (2006). The literature indicated that the self-efficacy dimensions like effort and persistence predict the risk-taking and creativity features of social entrepreneurs (Rahman & Pihie, 2014). Nonetheless, while social entrepreneurs are less confident in their abilities than their conventional counterparts, they are more likely to believe that they have the necessary skills compared to general adult population (Harding & Cowling, 2006). As a result, there is some evidence in the literature to support the idea that a person’s S-ENT intention is influenced by how they view their own capabilities (Baierl et al., 2014). Consequently, it can be hypothesized that:

H3:

The likelihood of someone becoming a social entrepreneur is favorably increased by that person’s confidence in his or her social entrepreneurial skills and knowledge.

1.6. Fear of failure (risk-perception)

The literature on entrepreneurial cognition has emphasized the importance of an individual’s risk tolerance (risk perceptions) in entrepreneurial decisions (Iyigun & Owen, 1998). Risk perceptions, on the other hand, are thought to be the result of failure fear (Arenius & Minniti, 2005), or fear of the undetermined financial or even social and psychological incentives that come with the process of venture creation. The literature demonstrated a significant negative relationship between fear of failure and the decision to start a new venture (Anjum et al., 2021; Arenius & Minniti, 2005; Langowitz & Minniti, 2007).

There is scarcity of research detailing how social entrepreneurs’ risk-taking and entrepreneurial activity interact, however, new research indicated that social entrepreneurs have lower levels of failure fear than the general adult population (Harding & Cowling, 2006). The literature lacks a comparison of social entrepreneurs’ propensity for risk-taking with that of their commercial counterparts. The decision to create a social enterprise is perhaps a difficult one for an individual to take, because it entails facing unusual difficulties (as described in introduction section of this document). Social entrepreneurs occasionally need to create value chains, access to markets, or even entirely new markets. Alternatively, they are tackling some of the most difficult challenges in the world, since they must reach far more people while spending lesser money (Bornstein, 2004). Risk-taking, therefore, is supported by the limited evidence as a crucial component of S-ENT orientation, both at the person and firm levels (Kraus et al., 2017; Satar & Natasha, 2019). Therefore, the following may be hypothesized:

H4:

Individuals who perceive greater degrees of fear of failure are less likely to become a social entrepreneur.

1.7. Perceived social passion

Social psychologists categorize passion as a motivational factor that might be behavioral, emotional, or cognitive (Chen et al., 2009). The passion is deeply established in the entrepreneurial endeavors (Cardon et al., 2017). Although entrepreneurs face substantial difficulties during the venture formation phase, such as a negative reaction to ideas, etc. (Cardon et al., 2009). Their passion allows them to persevere in the face of adversities, making it more pertinent for the start-up phase (Cardon et al., 2017). Additionally, researchers have shown that entrepreneurial passion increases an entrepreneur’s self-confidence among those who are not yet actively or legally involved in entrepreneurship (De Mol et al., 2020). Accordingly, entrepreneurship research has demonstrated that entrepreneurship passion can predict both the intention to start a business and its overall success (Karimi, 2020).

Likewise, social entrepreneurs are strongly motivated by a compelling social vision that encompasses a tenacious commitment and ardor for addressing a social issue (Barendsen & Gardner, 2004). As a result, social entrepreneurs are described as “those with a focused emphasis on venture success” and “people with a passion for tackling social concerns” (Satar & Natasha, 2019; Sharir & Lerner, 2006). Through factors including vision, goals, competency, motivation, self-efficacy, competitive strategy, and social passion, it has been connected to the creation and growth of social enterprises (Barendsen & Gardner, 2004). So, in the context of S-ENT, social passion refers to a powerful, positive emotion to pursue a vision of generating social value through the creation of a social enterprise (Barendsen & Gardner, 2004). The commitment to social issues is frequently complemented by feelings of emotional attachment as well as a sense of obligation to support a social or environmental cause (Keogh and Polonsky, 1998). The social vision pushes the entrepreneur’s ability to identify current opportunities objectively on the path of becoming agents of social change (Barendsen & Gardner, 2004). Emerging research showed that social passion and its constructs are a vital component of social entrepreneurs’ entrepreneurial orientation (Satar & Natasha, 2019) and has a role in explaining SEI (Bornstein, 2004; Cardon et al., 2017; Satar & Natasha, 2019). This is supported by a recent study of Chandra et al. (2021), which contended that EP may play a role in predicting individuals’ SEI. Similar to this, it has lately been asserted that emerging social entrepreneurs need to have a strong prosocial orientation in order to succeed (Ganguli et al., 2021).

While passion influences human behavior (Murnieks et al., 2014), the desire to start a new business comes first (Kolvereid, 1996). Because the idea of starting a new business precedes the start-up or planning phase (Chen et al., 2009), the entrepreneur’s founder role influences the entrepreneurial intention (for example, see Nasiru et al., 2015). In our setting, an individual’s social passion motivates the individual to recognize social issues and look for creative solutions. To indicate how ardent social entrepreneurs are about resolving a social issue or producing social value, we use the phrase “perceived passion.” Although little is known about the function of social passion in S-ENT, preceding discussion offers some insights for the following hypotheses:

H5:

Individuals who perceive greater degrees of social passion are more likely to become social entrepreneurs.

Based on the above discussing and considering the gaps in the body of knowledge identified in the discussions regarding the hypothesis for each of the construct, and also considering the controlling variables and absence of studies for identifying the impact of perceptual beliefs of the potential social entrepreneurs, the following research framework has been developed as mentioned in Figure 1:

Figure 1. Proposed conceptual framework

2. Methodology

2.1. Respondent selection

The study aimed at examining the influence of individual perceptions on the likelihood of becoming the social entrepreneurs. As per the GEM methodology of S-ENT surveys, the identifying factor for social entrepreneurs is an implicit mention of the social mission (Lepoutre et al., 2013; Terjesen et al., 2012). Accordingly, the current survey identified individuals as potential social entrepreneurs based on their responses to the screening question, “Are you, alone or with others, attempting to start any kind of activity, organization, or initiative with social, environmental, or community objective within the next 3 years?” (Terjesen et al., 2012). Furthermore, because our target population is the potential social entrepreneurs, we excluded everyone, active at any level of the social entrepreneurial process (including nascent and established social entrepreneurs). A total of 2710 individuals were identified qualifying the above criteria.

3. Survey administration

A questionnaire survey was administered between June 2020 and March 2022 among the adult population in the three Gulf Cooperation Council (GCC) countries (Saudi Arabia, Bahrain, and the United Arab Emirates (UAE)). The rationale for selecting these countries is that their current level of entrepreneurship understanding is better or equal to the MENA (Middle East and North Africa) region (Miniaoui & Schiliro, 2016). Data collection took a long time because of the COVID-19; however, the respondents responded and common method bias test was also conducted; however, there was no significant difference in the responses of the respondents. The survey was conducted using online through the Google Forms due to the outbreak of Covid-19, hence the data was collected online. The survey was administered to a sample of the adult population in each country (18–64 years old). It resulted in a cross-country total of 2868 people. After removing the individual-level missing data, the final sample included only 2632 respondents across the three countries. The size of the sample varied from 940 individuals in UAE to 632 in Bahrain and remaining 1060 from Saudi Arabia.

3.1. Variables

3.1.1. Dependent variable

The likelihood of becoming social entrepreneur was measured by whether the respondents were thinking about starting any kind of activity, organization, or initiative with a social, environmental, or community goal within the next 3 years. This variable has already been utilized in other studies including GEM survey (see Lepoutre et al., 2013; Terjesen et al., 2012).

3.1.2. Independent variables

The independent variables are the five perceptual variables of the study i-e., confidence in one’s skills and knowledge (confidn), ability to perceive opportunities (opportpercep), knowing other social entrepreneurs (knowent), fear of failure (fearfail), and perceived social passion (perpass). Following previous studies (Elsevr1; Arenius & Minniti, 2005; Camelo-Ordaz et al., 2016; Langowitz & Minniti, 2007; Terjesen et al., 2012), the above independent variables were measured using the binary variables. However, the measurement question for “perceived social passion” was synthesized from the extant literature in order to be consistent with the binary scale. All the five independent variables are binomial with possible answers “Yes” or “No” [Coded as 1 if the answer is “Yes” or 0 otherwise]. Although the construct validity of the single-item measure has received some criticism (Hayton et al., 2013, for example), the binary scales for the perceptual variables are employed in the GEM survey as well as other related studies.

The “confidence in one’s skills and knowledge” was assessed through the individual responses to the inquiry of whether respondents believed they had the knowledge, skill, and experience required to launch a new social business. Regarding the ability to recognize opportunities, we evaluated respondents’ awareness of the possibilities for launching a social enterprise in his or her area during the following six months. Similarly, whether the respondent personally knows someone who has launched a social business in the previous two years was used to determine whether they “know a social entrepreneur.” Fear of failure was evaluated through individual answers to the question of whether respondents identify this perceptual factor as a barrier to the creation of social enterprise. Finally, perceived social passion was determined if respondents felt strongly about the social issue or the S-ENT goal of their intended social enterprise. For all variables, the responses “I don’t know” or “respondent refused to answer the question” were coded as [8].

Additionally, the analysis established alternative models for evaluating the probability of dependent variable with the following control variables: age, gender, work status, education, household income, and the country. The (Table 1) presents the elaboration of selected control variables. This set of control variables have been found as pertinent in informing for entrepreneurial behavior studies (Lepoutre et al., 2013; Terjesen et al., 2012). IBM SPSS Statistics for Windows (SPSS Statistics 28.0) was used for data analysis.

Table 1. Description of control variables

Variable	Respondent question	Outcome(s) format	Nature of variable
Age (age)	Provide year of birth.	Continuous/absolute number	Continuous
Gender (gen)	Provide gender.	Male = 1 Female = 0	Categorical
Work Status (workstat)	Provide occupational status	Full or part time work Not working’ Retired or student	Categorical
Education (edu)	Provide the highest educational degree	No education Some secondary education Secondary degree Post-secondary education Graduate degree	Categorical
Household income (houseinc)	Provide information of household income	Upper income group 33% tile Middle income group 33% tile Lower income group 33% tile	Categorical
Country	Provide country of residence	If “Country X” = 1, if “other” = 0 Reference country: UAE	Dummy

4. Data analysis

Before starting the analysis it is important to be sure that the data is normal. Another important thing is to know the descriptive of the demographic and control variables. The descriptive of demographic variables and also the descriptives of the variables of the study including the independent and dependent variables are mentioned in Table 2 below:

Table 2. Descriptive statistics

Variable	Mean	Standard deviation	Skewness	Kurtosis
Social Ent	0.91	0.031	0.09	0.03
Confidn	0.87	0.045	0.04	0.01
Fearfail	0.83	0.037	0.10	0.06
Knowent	0.75	0.026	0.06	0.02
Opprtpercep	0.84	0.065	0.07	0.03
Perpass	0.88	0.049	0.06	0.07
Age (age)	38	18	1.2	1.04
Gender (gen)
Males (1179)
Females (1453)
Work Status (workstat)	2.9	1.13	0.10	0.31
Education (edu)	3.2	1.51	0.02	0.19
Household income (houseinc)	3	1.27	0.05	0.21
Country	0.60	0.037	0.06	0.01

A correlation matrix of the variables was computed (Table 3). According to the results of correlation analysis in Table 3, the predictor variables show correlation (p < 0.01) with the measure of being a social entrepreneur. The possibility of becoming a social entrepreneur reduces with age (r = −0.056). The females are seen to be more inclined towards being social entrepreneurs than male respondents (r = −0.075). Similarly, working individuals are more expected to qualify being as social entrepreneurs as compared to individuals in other occupational groups (r = 0.045). Likewise, the control variables of “education” has shown a positive while being the “household income” has shown a negative association with the dependent variable (r = 0.40 and −0.033 respectively). Finally, the effects of perceptual variables are as expected. The “confidence in one’s skills and knowledge” (r = 0.156), “knowing other social entrepreneurs” (r = 0.107), “ability to perceive opportunities” (r = 0.098), and “perceived social passion” (r = 0.096) are positively correlated to the measure of the dependent variable, whereas fear of failure (r = −0.059) is negatively associated with our social entrepreneur measure. There is no collinearity reported among the independent and control variables.

Table 3. Correlation matrix

		1	2	3	4	5	6	7	8	9	10	11
1	SocialEnt	1
2	Confidn	0.156	1
3	Fearfail	−0.059	−0.110	1
4	Knowent	0.107	0.253	−0.026	1
5	Opprtpercep	0.098	0.217	−0.082	0.217	1
6	Perpass	0.096	−0.028	−0.092	−0.068	−0.108	1
7	Age	−0.056	−0.116	−0.060	0.234	−0.153	−0.066	1
8	Gender	−0.075	−0.020	0.106	0.141	0.064	−0.121	0.022	1
9	WorkStat	0.045	0.038	0.103	0.096	0.080	0.041	−0.022	−0.203	1
10	Educt	0.040	0.085	−0.036	0.087	−0.061	0.098	−0.088	−0.078	0.072	1
11	Houseincm	−0.033	0.123	−0.031	0.107	0.082	0.068	−0.144	0.028	−0.041	0.056	1

5. Logistic regression

The binomial nature of dependent, as well as the independent variables, justified the use of binomial logistic regression for analysis (Hosmer et al., 2013), which assesses the probability of an event happening. The event (dependent variable) in the current research is the likelihood to become a social entrepreneur. The logistic regression will enable the comparison of different perceptions as well as help in establishing their individual and overall impact on becoming a social entrepreneur. Logit coefficients that indicate the changes in the log-odds of the dependent variable were calculated using maximum likelihood estimation. To simplify the understanding of how the probability of dependent variable changes with the different variable, we converted the odds ratio into probability by the using the equation: p=odds/1+odds; where “p” is the probability of the dependent variable and “odds” is the odds ratio of the variable being tested. The models’ goodness-of-fit was evaluated using Nagelkerke’s R-squared rather than Cox and Snell’s R-squared because the latter is based on sample size, log-likelihood of the model, and the original model, which never reaches its theoretical maximum of 1 (Field, 2009). Further, Wald statistics were also used to evaluate the significance of independent variables.

6. Models

A total of 7 binomial regression models were estimated for the variables. Model 1 is the base model which examines the link between basic demographic and economic characteristics (control variables) and the likelihood of being a social entrepreneur (dependent variable). The second model (Model 2) is used to test the first independent variable of “confidence in one’s skills and knowledge” and includes the control variables as well. Similarly, Model 3, Model 4, Model 5, and Model 6 investigate, respectively, the independent variables of “fear of failure”, “knowing other social entrepreneurs”, “ability to perceive opportunities”, and “perceived social passion” along with the control variables. The model 7 incorporates all the independent and control variables to investigate the combined effect on dependent variable. Finally, the country dummies are included (but not shown in the regression table). All models are built using data from all the countries surveyed (both men and women).

7. Results and discussion

The regression analysis results are demonstrated in Table 4. At the outset, the demographic variables were entered in Model 1. The results indicate that the individual’s demographic makeup is contributing to his/her likelihood of being a social entrepreneur. The overall model is significant at the p ≤ 0.001 level and correctly predicts 95 percent of the responses. The variable “age” has a significantly negative impression on the dependent variable. The coefficients of age (odds ratio = 0.818) show a significant negative relationship with the probability of being a social entrepreneur in all the models. This is consistent with existing evidence in the entrepreneurship literature, in which the likelihood of venturing is declared to be high at a young age and then declines (Levesque & Minniti, 2006). The age has been seen as affecting how the entrepreneur uses and activates his/her social networks which in turn affects their ability to obtain entrepreneur resources from social networks (e.g.: Greve & Salaff, 2003). Interestingly, the variable of “gender” demonstrates a significant negative relationship (odds ratio = 0.906 with a 95 per cent confidence interval). This suggests that in case of the S-ENT, women are relatively more likely to start a new business than males. This proves the claim that S-ENT activities are expected to be greater among women than men (Hechavarria et al., 2017). Further, the gap in gender is small in S-ENT as compared to conventional entrepreneurship venturing (Teasdale et al., 2011). Despite the fact that female entrepreneurs have different social networks than male entrepreneurs (Runyan et al., 2006), the current finding of women being comparatively more likely to start a new business in the S-ENT context warrants further investigation. The “work status” variable’s results (odds ratios = 1.610) show a significant positive relationship with the dependent variable. As a result, working people are 61 percent more likely to become social entrepreneurs than those who are “not working” or belong to any other occupational group. Furthermore, when people move one unit from not working to working status, their chances of becoming a social entrepreneur increase by 0.616 points. This assumes that many people start new social businesses while still working full-time (Arenius & Minniti, 2005). Education is also associated with a higher likelihood of becoming a social entrepreneur (odds ratios = 1.894). Indeed, as education levels rise, so does the likelihood of becoming a social entrepreneur. Previous research has linked higher levels of human capital to a higher likelihood of entrepreneurial activity (Davidsson & Honig, 2003).

Table 4. Logistic regression results: Dependent variable = (likelihood of becoming a social entrepreneur)

	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7
	B	Wald	B	Wald	B	Wald	B	Wald	B	Wald	B	Wald	B	Wald
	(S.E)		(S.E)		(S.E)		(S.E)		(S.E)		(S.E)		(S.E)
Perceptual Variables
Confidence in one’s skills & knowledge			1.389***	518.484									1.276***	484.000
			(0.061)										(0.058)
Fear of failure					−0.678***	66.585							−0.587***	80.564
					(0.083)								(0.062)
Knowing other Entrepreneurs							0.706***	225.619					0.711***	173.166
							(0.047)						(0.054)
Ability to perceive opportunities									0.589***	110.542			0.647***	148.932
									(0.056)				(0.053)
Perceived social passion											1.083***	509.010	1.083***	192.718
											(0.048)		(0.078)
Control Variables
Age	−0.202***	9.784	−0.302***	14.231	−0.206***	15.693	−0.021***	17.64	−0.013***	10.562	−0.022***	1.344	−0.161***	13.355
	(0.064)		(0.080)		(0.052)		(0.005)		(0.004)		(0.006)		(0.044)
Gender	−0.098***	9.392	−0.271***	29.376	−0.131***	4.759	−0.437***	18.355	−0.266***	21.777	−0.346***	10.456	−0.201***	1.467
	(0.032)		(0.050)		(0.060)		(0.102)		(0.057)		(0.107)		(0.137)
Work Status	0.478***	29.504	0.348***	41.57	0.222***	48.128	0.015***	1.562	0.0872***	1.745	0.3311***	13.238	0.401***	35.821
	(0.088)		(0.054)		(0.0320)		(0.012)		(0.066)		(0.091)		(0.067)
Education	0.639***	139.988	0.248***	363.796	0.306***	44.231	0.540***	50.447	0.462***	10.437	0.088***	1.832	0.456***	117.167
	(0.054)		(0.013)		(0.046)		(0.076)		(0.143)		(0.065)		(0.042)
Household income	0.031***	106.743	−0.134	124.62.	−0.154	92.590	0.107	19.839	0.459	205.743	−0.088***	21.445	−0.480***	199.211
	(0.003)		(0.012)		(0.016)		(0.021)		(0.032)		(0.019)		(0.034)
Model diagnostics constant	−1.778		−1.188		−4.894		−3.575		−2.552		−3.087		−2.731
	(0.096)		(0.113)		(0.169)		(0.166)		(0.120)		(0.308)		(0.132)
N	2632		2589		2601		2579		2583		2601		2612
Overall % correct predictions	95.0		95.0		94.8		95.0		94.9		95		95
R^2	0.090		0.104		0.161		0.116		0.105		0.160		0.167

*** significant at p ≤ 0.001; ** significant at p ≤ 0.01; *significant at ≤ 0.05. B= Coefficient; S.E = Standard Error

Despite the fact that the control variable “household income” is positively associated with the dependent variable, it has an odds ratio of (1.01), which is very close to 1. If we approximate it to be equal to 1, we could argue that there is no difference between the probability of being a social entrepreneur and the individual’s “household income”. This is an unexpected result and contrasts with the prior research in entrepreneurship behaviour (Arenius & Minniti, 2005). For the possible explanation, we could retrieve from existing literature that higher levels of income reduce the individuals’ financial barriers while at low levels of household income the individuals are prosperous of employment (Arenius & Minniti, 2005). This resonates with the recent evidence in literature that S-ENT activities are becoming prevalent among university students worldwide (Shahid & Alarifi, 2021) due to high unemployment levels (Alsaaty et al., 2014).

Finally, a dummy for each country was created (e.g., for UAE = 1, otherwise 0), and UAE was chosen as the reference country (coded as 1 on all other country dummies). The UAE was chosen since its rate of budding social entrepreneurs is 4.6 percent, which is near to the average of the three countries in our sample. For all the three countries, the country effect dummies are significant. Gender and age continue to have significant and negative effects. Furthermore, both working as well as non-working people are equally expected to engage in S-ENT. The effects of post-secondary and graduate education disappear. All of the perceptual variables continue to be significant. Except for “fear of failure,” all of the perceptual variables’ coefficients remain positive and significant. The fear of failure has a negative and statistically significant relationship with the dependent variable. As a result, the introduction of country-specific effects, even though significant, does not diminish the cross-country significance of perceptual variables. Saudi Arabia’s odds ratio is less than one. As a result, individuals in this country are less likely to become social entrepreneurs than in the reference country of the UAE. This implies that local influences play a role. The existing literature also supports the notion that the macroeconomic environment of some countries encourages entrepreneurship while penalizing it in others (Reynolds et al., 2003). Nonetheless, there is a link between cross-country and country-specific factors of social entrepreneurship behaviour, which is a complex issue that requires additional research.

8. Regression models: Perceptual variables

The perceptual variable “confidence in one’s skills and knowledge” is added to the demographic variables in Model 2. The importance of demographic variables is nearly unchanged in this model. The predisposition to be a social entrepreneur is positively and significantly associated with confidence in skills and knowledge. The variable has Wald statistics of 518.484 (significant at the 0.01 level) and an odds ratio of 4.010 with a 95 percent confidence interval. Individuals who are confident in their knowledge and skills are 301 percent more probable to become social entrepreneurs than those who are not, according to these statistics. A one-unit increase in confidence increases an individual’s likelihood of becoming a social entrepreneur by 0.800 points. The “confidence in one’s abilities” appears to be the most important aspect of the S-ENT intention. This variable’s positive and strong influence corresponds to evidence in existing research in psychology, self-efficacy, social, and entrepreneurship literature (Harding & Cowling, 2006; Hockerts, 2015). Model 3 examines the variable “fear of failure.” The variable has a significant negative influence on being a social entrepreneur (odds ratio = 0.4588). This is consistent with the entrepreneurship literature, which states that an increased perception of the likelihood of failure raises the perceived riskiness of starting a business (Weber & Milliman, 1997). Risk-taking has been demonstrated to be a valid consideration in the study of S-ENT orientation at both the individual and firm levels (Harding & Cowling, 2006; Kraus et al., 2017; Satar & Natasha, 2019). Model 4 examines the variable “knowing other social entrepreneurs.” The variable has a significant positive relationship with the likelihood of becoming a social entrepreneur (odds ratio = 2.025). Existing research has shown that the influence of role models or being a part of networks can reduce ambiguity and provide resources for new entrepreneurs (Weber & Milliman, 1997). Nonetheless, the findings make no distinction between positive and negative role models or examples. According to recent literature, the influence of knowing other social entrepreneurs, social capital, trust-building, and connection links are seen as important characteristics of social entrepreneurs (Satar, 2019; Satar & John, 2019). Model 5 assesses the “ability to perceive opportunities.” The variable has a positive and statistically significant relationship with S-ENT intention. The odds ratio for the variable is 1.805 which leads to an 80.5 per cent higher likelihood of an individual to become social entrepreneur when perceiving opportunities. This corresponds to the literature where alertness to unexploited opportunities is a precondition for both social and commercial entrepreneurship venturing (Urban, 2020).

Finally, in Model 6, the “perceived social passion” is positively and significantly related to the likelihood of becoming a social entrepreneur. According to the variable’s odds ratio of 2.952, individuals who are passionate about their S-ENT mission are 195 percent more likely to become social entrepreneurs than those who do not perceive such passion. The recent literature has started exploring the role of entrepreneurship passion in the venture formation stage (Cardon et al., 2009), opportunity perception (Murnieks et al., 2014), building self-confidence and influencing the entrepreneurship intention in entrepreneurs (De Mol et al., 2020; Karimi, 2020). Indeed, some relevance can be made with the anecdotal evidence of studies like Lurtz and Kreutzer (2017) and Satar and Natasha (2019) who have attempted to explore the role of passion in S-ENT.

Finally, in Model 7 overall effect of five perceptual variables is tested for S-ENT intention. The “confidence in one’s skills and knowledge” has an odds ratio of 3.586 which is slightly lower than its odds score in Model-2. Nevertheless, the variable has a higher odds ratio than all other perceptual variables. Although the “fear of failure” has a slightly increased odds ratio of 0.555, it still maintains a negative relationship. Similarly, someone who knows other social entrepreneurs is 103.60 percent more likely to become a social entrepreneur than someone who does not know other social entrepreneurs (the odds ratio of knowledge is −2.036, as in Model 4). In a similar vein, being “able to perceive opportunities” increases an individual’s likelihood of becoming a social entrepreneur by 90 per cent. (Odds ratio in Model 7 is 1.901 slightly higher than odds ratio in Model 5). Finally, perceived passion remains statistically significant with an odds ratio of 2.958, which is similar to its score in Model 6. Individuals who feel passionate about a social problem are thus 195 per cent more likely to become social entrepreneurs.

In addition, the Nagelkerke R-squared results reveal that the base model (Model 1) has the lowest R-squared value. All other models’ R-squared values increase when compared to the base model, confirming that adding a new variable to the base model improves the new model’s explanatory power. Model 7 has the maximum R-squared (R = 0.167), indicating that the variation of the independent variables explains 16.7 percent of the variation of the dependent variable. As a result, including all perceptual variables in Model 7 adds value to the model and improves its explanatory power. In total, the perceptual variables were identified as potential predictors of S-ENT intention by the analysis. Indeed, the influence of perceptual variables may be greater than the influence of demographic control variables (Model 1).

9. Conclusion

According to the entrepreneurship literature, perceptual factors play a significant role in a person’s propensity to engage in entrepreneurial activity (Arenius & Minniti, 2005). However, the impact of perceptions on social entrepreneurial intentions is either absent or fragmented in the literature. Therefore, the present work was formulated with the empirical goal of identifying and analyzing the individual perceptual factors influencing the likelihood of becoming a social entrepreneur. Following an extant review of literature, the study analyzed the adult population data from the questionnaire survey administered across three GCC countries. The original data was used to create binomial logistic regression models considering the economic and demographic variables. Subsequently, the five identified perceptual variables were later added to the regression models. The addition of perceptual variables improved the model’s statistical fit. The macroeconomic differences were eventually accounted for by the country dummies.

The analysis revealed that the fundamental subjective factors of “confidence in one’s skills and knowledge,” “knowing other social entrepreneurs,” “ability to perceive opportunities” and the “perceived social passion” are the main individual perceptions influencing the likelihood of becoming a social entrepreneur. This holds good for all the potential social entrepreneurs across all the studied countries and genders. Among the analyzed perceptual variables, the “confidence in one’s skills and knowledge” has a key positive impact on the S-ENT intention. Our findings also show that “perceived social passion” “knowing other social entrepreneurs,” and “ability to perceive opportunities” collectively shape the S-ENT intention, while the “fear of failure” reduces the chances of becoming a social entrepreneur. Specifically, the study found that entrepreneurial self-efficacy, role model perception, and opportunity perception play similar roles in traditional and S-ENT intention formation, albeit to varying degrees of influence. The negative effect of “fear of failure” in the context of S-ENT, on the other hand, is intriguing due to the widespread belief that S-ENT is less risky than commercial alternatives. One argument is that because S-ENT targets dual or triple bottom line goals, it should be deemed equally or more risky than commercial entrepreneurship. Second, to explore the role of the macroeconomy, the study remarkably considered the country-specific effects as well. As a result, the work may allow for descriptions of S-ENT behavior in which intention formation is based not only on objective demo-economic issues, but also on subjective individual perceptions. To summarize, our findings broaden the cognitive approach to S-ENT behavior and shed light on the role of perceptual variables in the S-ENT antecedents. As such, our research contributes to the economic theory of entrepreneurial motivation and, more specifically, to our understanding of SEI.

10. Limitations and future research

10.1. Limitations

The study pertains to a survey of the adult population in only three GCC countries. Since S-ENT is deeply embedded in socio-economic environments, the role of context needs further validation. Second, the list of social entrepreneurial intentions and perceptions elements covered in this work is not exhaustive because social entrepreneurs offer enormous possibility for varied perspectives surrounding the cognitive and perceptual aspects of their S-ENT intention. Third, the nature of the questionnaire items precludes the application of more precise statistical techniques, such as the structural equation models, which could reveal the various links between perceptions and intentions. Therefore, the absence of major theoretical variables should be attributed to the models’ minimal variance.

We believe that the multinational data covered in the present study is an excellent starting point for investigating these cognitive aspects of the S-ENT perceptions. However, if a more in-depth investigation of the potential social entrepreneurs is to be conducted, the survey will need to be supplemented with added items and other changes. Finally, but certainly not least, the project was partially implemented during the COVID-19 pandemic. We acknowledge that the Corona pandemic, as well as the individual and social restrictions that followed, may have influenced people’s attitudes, perceptions, and behaviors toward business and entrepreneurship in general. This might have favorably or unfavorably influenced the individual perceptions towards S-ENT venturing.

11. Scope for future research

This is the distinctive study to use general population survey data to understand S-ENT perceptions. In this regard, the results were especially convincing because they confirmed a priori assumptions. More research, however, is required to confirm or refute these findings. Additional assumptions should be investigated using diverse samples to improve understanding in this area.

With huge diversity manifested in social enterprise structure, strategies, processes, and missions, the behavior of social entrepreneurs is still fascinating ground to researchers worldwide. As such, it offers immense potential for diverse perspectives and literature for discovering the unique and intricate complexities surrounding the cognitive and perceptual decision-making of social entrepreneurs. Although our project is grounded on extant theoretical and empirical investigation, it would be appropriate to design more in-depth analytical approaches to reveal how, and why individuals choose to venture out in S-ENT and how they are exposed to and influenced by different perceptual factors in their venturing decisions. Meanwhile, adding more variables may allow for a more comprehensive assessment of the relative likelihoods of expressing entrepreneurial intentions for many distinct categories of respondents, such as “women.” A new questionnaire, on the other hand, could be developed and evaluated to address some of the shortcomings of the one used here.

Additionally, although we could find evidence for ‘fear of failure; with the proxy of “risk taking” as contributing to social entrepreneurial intention, it needs further examination. Similarly, the pioneering findings of social passion as well its interaction with perceptual variables and decision-making in S-ENT venturing demand further probes.

Furthermore, the data for considering country-specific effects were restricted to the inclusion of only dummy variables. Thus, it would be more appropriate not only to specify the cross-country differences but also to analyze the role played by macro-economic conditions like institutions and culture, level of unemployment, technological development, or other macro factors of the S-ENT ecosystem.

Disclosure statement

No potential conflict of interest was reported by the author(s).