The Manifestation of Entrepreneurial Orientation in the Social Entrepreneurship Context

Abstract

This study extends theorisation of entrepreneurial orientation (EO) in the social entrepreneurship context by introducing firm performance as a construct for examining social enterprises (SEs) EO. Drawing on EO and firm performance research indicating that EO is related to a better firm performance, this paper argues that EO in SEs is positively related to organisational performance. This research empirically studies 303 social enterprises in Saudi Arabia and develops three hypotheses that examine the relationship between entrepreneurial orientation (EO) and social entrepreneurs’ firm performance. The results show that innovativeness and proactiveness, but not risk-taking, are positively associated with firm performance. Thus, whilst SEs cover a wide range of business activities there is generally a positive effect of EO across the contexts investigated. Also, this research found that EO can be used as a mechanism to overcome constraints imposed by limited resources in an environment where new opportunities rarely occur.

Keywords:

• Entrepreneurial orientation
• social enterprises
• social entrepreneurship
• Saudi Arabia
• firm performance

Introduction

Social entrepreneurship (SE) is considered an emerging field of study (Nicolás, Rubio, and Fernández-Laviada 2018) that has attracted enormous attention. Entrepreneurial orientation (EO) on the other hand is a prominent, widely used theoretical construct in management research (Covin and Lumpkin 2011; Rauch et al. 2009; Wales, Gupta, and Mousa 2013). In SE research to date, researchers have been reluctant in using EO. Although the concepts seem naturally connected, there is little research bridging the relationships between EO and SE. A reason for this could be, a lack of operationalising entrepreneurial attitudes and behaviors in the SE context due to the priority of the social mission over the economical one (Lurtz and Kreutzer 2017). Another reason is that entrepreneurship logic is not that clear in this context due to the multiple stakeholders involved with the enterprise (Morris et al. 2007). While much of what is know about SE is based on the related area of corporate entrepreneurship (Griffiths, Gundry, and Kickul 2013). Research suggests that the existence of EO, which consists of innovativeness, proactiveness and risk-taking is what makes an entrepreneur entrepreneurial (Anderson et al. 2015). It is about time to use corporate entrepreneurship theorisation of EO to push forward the understating of performance in the SE context. While researchers in strategic management and corporate entrepreneurship have produced considerable evidence that high EO leads to better firm performance (Rauch et al. 2009), and prior research has generated a wealth of insights about how EO manifested in entrepreneurial organisational, almost no attention has been devoted to this relationship in the SE context. This paper is establishing a foundation of the relationship of EO and SEs performs to build on by future research.

This paper seeks to make three sets of contributions. Firstly, this study is filling a gap in the literature by quantitatively investigating entrepreneurial orientation (EO) on SEs performance using a new data set. EO is associated with enhanced performance, though it is unclear if it affects SEs’ outcomes (Lumpkin et al. 2013). Secondly, this research responds to SE and EO researchers’ calls to use quantitative research focusing on developed countries. Doherty, Haugh, and Lyon (2014) have suggested that future research in SE should focus on developing countries including those in the Middle East. Wales, Gupta, and Mousa (2013) also encourage future EO research to focus on countries where EO is unexamined such as those in Middle East countries. Thirdly, this research responds to calls for understanding the influence of dimensions of EO on firm outcomes such as firm performance (Wales, Gupta, and Mousa 2013) and looking at the SEs’ roles in managing their ventures by taking competitive advantage actions. The literature suggests that EO effects performance positively in entrepreneurial firms and furthermore, this paper intends to check if EO can influence performance in this context. However, the literature is not conclusive if EO is usable without any modifications to its scale in the SE context and therefore, this study will test each EO dimension separately to test for positive effects with social enterprises’ performance. This leads to the research question: How do the individual dimensions of EO influence the social enterprises’ performance?

Theoretical background

The multidimensionality of SE makes it an interesting field for different perspectives (Mair and Marti 2006). The following sections adapt corporate entrepreneurship perspectives to the SE context and summarise the relationship between EO, and social enterprises’ performance that inform the derivation of these three research hypotheses.

Firm performance

Firm performance is considered one of the main constructs in management research (Richard et al. 2009); however, it has an inconspicuous meaning due to its complexional multidimensionality (Gupta and Wales 2017). While empirical studies measuring firm performance use a variety of indicators (Rauch et al. 2009) those indicators can be divided into subjective and objective measures (Brush and Vanderwerf 1992; Venkatraman and Ramanujam 1986). Richard et al. (2009) suggest that using subjective measures allows for the assessment of nonfinancial standards of performance. Frank and Roessl (2015) further explain that the popularity of subjective measures is due to the difficulty of obtaining objective indicators, especially in small and medium-size enterprises.

Studies in an international context have revealed that subjective data sources are a credible measurement of performance. For example, Lukas, Tan, and Hult (2001) mentioned that in the context of emerging economies like China, subjective measurement can be a more reliable source than objective one for measuring performance. Likewise, Brouthers (2002) explained that researchers should use a subjective measure of performance when they not only aim to understand the goals of a particular strategy but also want to gain insight into managers’ views about performance management goals. In this research, subjective measures of performance are used because of the sensitivity of reporting confidential financial information (not publicly listed). Furthermore, the use of financial indicators in the SE context might not be the best way to measure performance because the social objective is superior to the profit maximisation objective.

Firm performance in the SE context

SEs strive to create a sustainable venture by acquiring and managing resources effectively and building the capabilities of their venture (Meyskens et al. 2010; Renko 2013). Hence, the focus is on the SEs ability of mission realisation through developing capabilities that are centered on serving their stakeholders (i.e. funders, donors, volunteers, employees, customers, beneficiaries). Therefore, SEs must attain multiple stakeholders either to serve them or to gain their support (Desa and Basu 2013). Serving such multiple stakeholders’ mission requires a subjective firm performance measurement (Bagnoli and Megali 2011; Schmidt et al. 2015).

Felício, Gonçalves, and da Conceição Gonçalves (2013) conducted a study to examine the influence of SEs and transformational leadership on performance and social value creation. The study revealed that SEs’ performance is reflected in their ‘ability to satisfy users, the quality of service and the success of organisation recognised by society’ (2144). Chen and Hsu (2013) used a subjective measure of performance in the SE context consisting of employee satisfaction, coordination among employees, satisfaction of service object and the prospect of organisation. Bacq and Eddleston (2016) proposed that in order for SEs to create larger social impact they should focus on three capabilities, ‘stakeholder engagement’, gain ‘government support’ and ‘developing revenue streams’. Bloom and Chatterji (2009) suggested seven capabilities for SEs to achieve their social goals. Other scholars used a goal directed approach to measuring SEs’ performance through three dimensions: economic performance, social effectiveness and institutional legitimacy (Bagnoli and Megali 2011; Schmidt et al. 2015). While others relied on measures of SEs’ effectiveness as an indicator of their performance (Chen and Hsu 2013; Knife, Haughton, and Dixon 2014; Miles, Verreynne, and Luke 2014) and other researchers suggest SEs’ performance measurements focusing on the SEs’ social and economic effectiveness (Arena, Azzone, and Bengo 2015; Arogyaswamy 2017; Miles et al. 2013).

The previous section shows that SEs’ performance research has emerged recently and it is considerably at a premature stage. To date, there are no unified measurement of SEs’ performance, and this is due to the multiple stakeholders of the SEs and their mission duality (social and economic). SEs should achieve their social objectives and be successful in building their capabilities by being resourceful to achieve their economic goals. Albert, Dean, and Baron (2016, 292) explain that ‘In the context of social entrepreneurship, “resource providers” are defined as individuals or organizations who offer support to the social venture’. Therefore, SEs seek the support of a wider stakeholder pool than their commercial counterparties (Moss et al. 2011). This study adapts the previous measure of SEs’ performance mainly relying on the SEs’ social and economic effectiveness.

Entrepreneurial orientation

EO is a widely used and accepted concept used in entrepreneurship research and has been adopted by other management disciplines. A definition of EO proposed by Miller (1983, 771) states that ‘an entrepreneurial firm is one that engages in product market innovations, undertakes rather some risky ventures and is first to design proactive innovations that beats competitors to the punch’. The work of Miller (1983) on firm’s strategy-making suggests innovativeness, pro-activeness and risk-taking as the three EO dimensions. Covin and Slevin (1989) developed a nine-item scale to measure the three dimensions of EO, which are: innovativeness, risk-taking and proactiveness. Covin and Wales (2010) and Miller (2011) further advocated that it is the combination of these dimensions in a person or organisation, which makes the person an entrepreneur and the organisation an entrepreneurial organisation. This research would use the EO conceptualisation developed by Miller (1983) and Covin and Slevin (1989) to investigate the entrepreneurialism of SEs.

Lumpkin and Dess (1996) added two dimensions competitive aggressiveness and autonomy to the original scale of EO. In their definition of EO, they tied the concept to the new entry process (i.e. ‘EO refers to the processes, practices, and decision-making activities that lead to new entry’ p. 136) and this may lead to the minimal use of this construct in comparison to Covin and Slevin’s (1989) conceptualisation of EO (Covin and Miller 2014). The literature suggests that SEs would not be aggressive nor competitive towards other players, on the contrary, they are collaborative as they all share a common goal of superior social benefit (Montgomery, Dacin, and Dacin 2012; Nicolás, Rubio, and Fernández-Laviada 2018). Covin and Miller (2014) further agree that competitive aggressiveness may not be evident in all contexts as some encourage collaboration in their entrepreneurial endeavor. Therefore, because these two dimensions are generally against the mission of SEs mentioned earlier in their definition, it has been decided not to use them in this study. On the other hand, focusing more on the three more widely used EO dimensions, we intend to provide a clearer and unbiased comparative analysis of SEs with commercial enterprises (CEs).

Lumpkin and Dess (1996) suggested that EO dimensions can be used independently as a multidimensional phenomenon instead of a unidimensional one. In both EO scales designed by Covin and Slevin (1989) and Lumpkin and Dess (1996) the three dimensions of innovativeness, pro-activeness and risk-taking are core in defining an entrepreneur. There is a debate in the literature on the multi-dimensionality of the EO construct, and Miller (2011) and Covin and Lumpkin (2011) ended the debate by acknowledging that the dimensions of EO can be examined separately. Wales, Gupta, and Mousa (2013) further support this notion by suggesting that the use of different dimensionality of EO should be based on the research question at hand. Both constructs can be used as suggested by Miller (2011) ‘in some research contexts, the best of both worlds may entail analyses that present results for the EO construct and for each of its components’ (880).

EO in the SE context

Most of what is known about EO in the SE context is based on nonprofit social organisations. Those studies can be divided into conceptual (Beekman, Steiner, and Wasserman 2012; Kusa 2016; Miles et al. 2013), qualitative (Lurtz and Kreutzer 2017; Syrjä et al. 2013) and quantitative research (Chen and Hsu 2013; Pearce, Fritz, and Davis 2010). Based on those research studies EO in social context is best characterised by the three dimensions of innovativeness, pro-activeness, and risk-taking. Furthermore, some researchers used EO scale without any modifications (Barrett, Balloun, and Weinstein 2005), while others used the EO scale with minor modifications (Morris et al. 2007). Lastly, studies that had major modifications to the EO scale by adding new items to the scale such as socialness, for example, have renamed the scale to Social Entrepreneurial Orientation (SEO) (Kraus et al. 2017).

Most of the studies on EO in the social context have measured either three or four dimensions, but innovativeness, proactiveness and risk-taking are central to most of the studies (DiVito and Bohnsack 2017; Kraus et al. 2017; Lurtz and Kreutzer 2017; Coombes et al. 2011; Syrjä et al. 2013). A clear-cut demarcation of EO in profit-earning SE and the non-profit sector may be hard to attain. Helm and Andersson (2010) noted that many social science concepts lack distinct boundaries with similar concepts. This obstacle has dominated non-profit entrepreneurship research. However, SE is a unique non-profit and profit behaviour existing at the meeting point of innovation, proactiveness and risk-taking, and there is a clear behavioural difference between non-entrepreneurial and entrepreneurial non-profits (Helm and Andersson 2010). Furthermore, there has not been a well-established modified scale of EO in the SE context. Therefore, in this research, Covin and Slevin’s (1989) EO scale will be used without any definitional changes.

EO and firm performance

EO and firm performance have been linked together increasingly in research publications (Gupta and Wales 2017) because of EO contribution to firm performance (Covin and Lumpkin 2011). Rauch et al. (2009) studied the relationship between organisational performance and EO. Their study indicated a strong relationship between firm performance and EO; the authors found that the strength of this relationship was influenced by moderator variables such as national culture, size of business and the extent to which the industry in which the firm operates is technology intensive. Lumpkin et al. (2013) emphasised that the outcome of EO should include elements of social value creation.

Academic literature has reported variations in the influence of EO on the performance of nonprofit SEs. While some studies have mentioned that EO influences performance positively (Pearce, Fritz, and Davis 2010), others have explained that market orientation influences it (Chen and Hsu 2013; Morris et al. 2007). Hu and Pang (2013) mentioned that social entrepreneurial orientation (SEO) influences a firm’s performance positively. Likewise, achieving the social objectives by creating social value contribute to the firm performance (Albert, Dean, and Baron 2016). Nonetheless, Coombes et al. (2011) found a negative relationship between EO and performance of nonprofit social enterprises. Thus, Lurtz and Kreutzer (2017) explained that in the context of nonprofits, the studies explaining EO’s influence on performance reported ‘mixed results’ (95). This discussion shows that while the relationship between EO and performance is positive, several factors—including environment, technology intensiveness of industry, national culture and sub-dimensions of EO—influence the relationship.

Derivation of hypotheses

EO is composed of three components, proactiveness, innovativeness and risk-taking (Covin and Slevin 1989). Proactiveness here is ‘an opportunity-seeking, forward-looking perspective involving the introduction of new products or services ahead of the competition and acting in anticipation of future demand to create change and shape the environment’ (Lumpkin and Dess 2001, 431). Proactive firms use information and knowledge to identify emerging opportunities and gain competitive advantage by investing in those opportunities; correspondingly, these firms might earn higher profits and brand recognition (Dess and Lumpkin 2005). In a dynamic environment, organisations can benefit from proactiveness as it allows them to gain first-mover advantage by responding to changes in the environment (Lumpkin and Dess 2001). Rauch et al. (2009) found a positive correlation between proactiveness and the performance of an organisation.

In the SE context, the ability of the SEs to be proactive is illustrated by the enactment of change on how social purpose is achieved and financial requirements are met relative to organisations with similar missions (Fairbourne, Gibson, and Dyer 2007). However, Austin, Stevenson, and Wei‐Skillern (2006) noted that SEs focus more on proactive activities rather than typical firm internal management activities. Tan and Yoo (2015) support that SEs are resourceful and accountable while pursuing their mission. Chen and Hsu (2013) sought to investigate if there exists an inverted U-shaped relationship between proactive behavior and SEs performance. The authors hypothesised that excessive proactiveness would hamper performance of SEs; however, this hypothesis was not supported. Kim, Lee, and Choi (2013) reported that proactiveness is positively related to financial performance of SEs.

SEs mainly depend on resources that are outside their organisation, unlike CEs (Gras and Lumpkin 2012). Thus, they seek a wider range of stakeholders for financial support (Austin, Stevenson, and Wei‐Skillern 2006; Newth and Woods 2014; Nicholls and Cho 2006). Social capital may often serve as a source of legitimacy that requires a commitment to stakeholders while not losing sight of the social mission (Nicolopoulou and Ozkan 2009) this type of proactive engagement with stakeholders further expand the SEs legitimacy and performance. Therefore, this study argues that the proactive EO dimension influences SEs’ performance positively. This leads to the first part of the first hypothesis of this study as follows:

Hypothesis 1: SEs’ proactive behavior is positively associated with SEs’ performance.

Risk taking implies that organisations will be better placed to invest resources in industries/markets, without being certain about the consequences of investment (Lumpkin and Dess 1996), thereby leading to an increase in generation of creative ideas (Wagener, Gorgievski, and Rijsdijk 2010) and long-run profit (Wiklund and Shepherd 2005). Rauch et al. (2009) found that risk-taking is positively associated with firm’s performance. Begley and Boyd (1987) found that risk-taking has a curvilinear relationship with the performance of entrepreneurial firms. To illustrate, the authors mentioned that firms with moderate risk-taking will perform better than organisations with very high or low levels of risk-taking.

Haughton (2008) observed that SEs have many times stepped in to meet the gaps of unsteadiness, where the governments have been unsuccessful, by highlighting social goals above financial returns. Among all the traits, the personality of social entrepreneur plays an important role in stimulating for taking risk. SEs are often differentiated by their ability to imagine, tackle, enable and present transformational changes efficiently while facing scarce resources, risks and diverse contexts (Thompson and Doherty 2006). Furthermore, the human capital may influence organizational performance and provide access to a wider range of opportunity (Cope, Jack, and Rose 2007).

SEs handle financial requirements differently; the willingness to take actions that have a positive social impact even if it possesses a magnitude of financial loss, loss in the amount of social impact incurred by the firm, and loss of non-financial stakeholder support, all point to the risk-taking tendencies of the SEs (Coombes et al. 2011). In their empirical research, Morris et al. (2007, 16) stated that ‘there may be important non-financial dimensions of risk’, which may be very difficult to quantify. The risk and return in monetary value are usually dependent on the social value and is accountable to different stockholders (Tan and Yoo 2015). Moreover, social missions are associated with rapid growth pressures that may involve greater financial risk-taking. However, addressing widely known social ills might require less risk-taking. On the downside, risk-taking jeopardizes the firm’s ability to address the social problems (Lumpkin et al. 2013). This study argues that the risk-taking dimension of EO has a noticeable impact on the performance of SEs. The second hypothesis of this study is:

Hypothesis 2: SEs’ risk-taking is positively associated with SEs’ performance.

Innovativeness has been positively associated with increased organizational performance (Zahra and Bogner 2000). Zahra (1996) explains that innovative behavior is crucial for determining the survival of firms, as, in the current era of competitiveness, firms have to use technologies to come with plans that allow them to show superior financial performance. SE authors such as Chell, Nicolopoulou, and Karataş-Özkan (2010) and Mair and Marti (2006) have noted that, due to the multidimensional origins of social problems, social entrepreneurs have various potential ways to exercise the tools or strategies of innovation to achieve their social mission. In particular, Alvord, Brown, and Letts (2004) noted that scarce resources could stimulate social entrepreneurs to think more creatively and to seek improved methods for tackling social issues, thereby producing high innovativeness. Alvord, Brown, and Letts (2004) argued that resource limitations lead to increased creativity among social entrepreneurs, resulting in more innovativeness within services and improvement in processes. This study might, therefore, consider innovativeness a significant factor in SEs conduct (Lepoutre et al. 2013; Lumpkin et al. 2013).

Coombes et al. (2011) propose that the emphasis be directed to the achievement of the SEs’ core mission, either by increasing efficiencies, serving more individuals or enhancing what is done for these individuals. In addition, the generation of new sources of revenue by the SEs, such as selling products or launching ventures that are supplementary to, or independent of the social mission, also depict innovativeness (Tracey and Jarvis 2007). SEs can also gain legitimacy through inter-organisational networks and strategic alliances (Hjorth 2013), such alliances are characterised by innovation and entrepreneurship, serving as places where knowledge creation and development are critical (Khoury and Pleggenkuhle-Miles 2011). Kim, Lee, and Choi (2013) investigated the factors that had an impact on economic and social performance of 185 social enterprises in Kenya. Correlational analysis was used to find a relationship among variables. The authors found that innovativeness had a positive effect on the economic performance of SEs. Since the literature implies that SEs are innovative, this research suggests that the innovativeness of EO dimensions influence firm performance positively. The third hypothesis of this study is as follows:

Hypothesis 3: SEs’ innovativeness is positively associated with SEs’ performance.

Methods

Researchers have used different methodological strategies to investigate the topics of SE, however, research in the field of SE is relatively new. Scholars have stated that there is limited high-quality quantitative research in the SE field (Doherty, Haugh, and Lyon 2014).

Sample

A comprehensive data list of SEs is not publicly available from a single source in Saudi Arabia. The researchers had to individually contact organisations in Saudi Arabia to assemble the sample source data base. The sources of data of firms for this research sample are combined from the following organisations: the Ministry of Labor and Social Development (MLSD), the King Khalid Foundation (KKF), the King Salman Youth Center and the Tasamy for Social Entrepreneurship. The previous organisations had lists of SEs working in the kingdom of Saudi Arabia due to working closely with them in, training, incubating, accelerating or grant competitions. The final Saudi Arabian sample size is a total of 1,870 social enterprises.

Data collection and respondents

An equivalent English and Arabic versions of the questionnaire were developed for this research. The process involved independent translations and back translations; pre-test reviews by three independent scholars, all of whom were bilingual in Arabic and English. A seven-point Likert scale was used in order to help reduce the statistical problems of extreme skewness (Matell and Jacoby 1971) and also to potentially result in a greater variance in the responses.

All the SEs were sent by email, an electronic survey through an online tool, ‘Qualtrics’, between 8 October 2016, and 15 December 2016. Follow-up emails and up to four reminders were sent after the initial invitation to take part in the survey. Of the 1,870 SEs sent emails to participate in this study, 683 respondents started the survey; however, only 350 respondents completed the survey, representing a response rate of 18.72%. The 350 responses included 35 respondents with silly/lazy responses (ticking the same box) and 12 respondents from large SEs. Those 47 respondents were discarded from the analyses; thus, 303 respondents remained for a response rate of 16.20%.

Self-reporting measures have been criticised for potential respondent bias; however, Rauch et al. (2009) did not find inflation between the firm performance and EO relationship due to self-reporting measures in their meta-analysis. Furthermore, a test for bias, following Bates and Creighton (2000), was used by dividing the respondents into two groups based upon whether the SEs completed the survey in response to the initial invitation, or responded after receiving one of the four reminders, assuming late respondents are somewhat similar to non-respondents. There were no statistically significant differences at the five percent level, or better, between the two groups against the social entrepreneurs’ gender, age, education, type of work (full time/part time/volunteer), or the SE characteristics such as age, size (measured as number of full time employees), and industry. The test was repeated by further dividing the respondents into 3 groups: those who responded to the initial email; secondly, those who replied to the first or second reminder; and a third group of those who replied to the third or fourth reminder. Again, there was no evidence of statistically significant differences at the five percent level, or better. Thus, the evidence suggests there is no systematic non-response bias (Barclay et al. 2002).

Measures

Dependent variable

The dependent variable (firm performance) was measured using 4 items. Respondents were asked, ‘How do you view your social enterprise? Please indicate the extent of agreeableness to each of the following statements by circling a number. Our organisation.’. This was followed by four statements, (i) ‘beneficiaries satisfied (clients, donors, staff and volunteers)’, (ii) ‘efficient in operations’, (iii) ‘attaining its stated goals and objectives’, and (iv) ‘able to adapt to the changing environment so as to attain its mission and vision during changing circumstances.’ Items adapted from the subjective scale developed by Miles et al. (2013), Chen and Hsu (2013) and Knife, Haughton, and Dixon (2014) for the subjective measures of social and economic performance in social enterprises.

Principal components analysis (PCA) was done on a 4-question questionnaire that measured resource acquisition on the 303 social enterprises (Table 1). Inspection of the correlation matrix showed that all variables had at least one correlation coefficient greater than r = 0.3. The overall Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was 0.767 with individual KMO measures all greater than 0.7. The PCA revealed one component had eigenvalues greater than one, which explained 65.42%of the total variance. In addition, the one-component solution met the interpretability criterion. As such, the component was retained. The four items Cronbach’s alpha (α  =  0.822) suggest a sound level of internal consistency.

Independent variables

EO was measured by following the 9 item 7 point scale which was developed by Covin and Slevin (1989). The EO scale items were divided into the three dimensions of innovativeness, proactiveness and risk-taking by adding the scores of each dimension and taking the average.

The EO scale used in this study was developed by Covin and Slevin (1989) and it is thus a previously tested, widely used and validated measure. Therefore, construct validity is not an issue as it mostly associated with newly established scales. This study adapted the Covin and Slevin (1989) scale without any changes to the nine items representing the three dimensions of proactiveness, innovativeness and risk-taking of EO. The EO scale in this study considered an acceptable reliability (α = 0.8) (Malhotra and Birks 2007). The overall Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was (0.824).

Control variables

Control variables relating to the social entrepreneurs were included. Six education dummy variables were considered as control variables (High School [no = 0, yes = 1], Diploma [no = 0, yes = 1], Bachelors Degree [no = 0, yes = 1], Masters Degree [no = 0, yes = 1], PhD [no = 0, yes = 1], and Other [no = 0, yes = 1]) (Harding 2006). The reference category in the models is a Masters Degree. This control variable was chosen because people with advanced levels of education are more likely to become social entrepreneurs, is consistent with Harding’s (2006) study. Male entrepreneurs were allocated a value of ‘1’ and female respondents were allocated a value of ‘0’ (Gender). Prior business experience is an important factor in creating social entrepreneurs (Van Ryzin et al. 2009) and therefore, respondents were asked, ‘Have you fully owned or partially owned a business in addition to this social enterprise?’ (Experience [no = 0, yes = 1]). Control variables relating to the social enterprises’ characteristics also were included. The natural logarithm of full-time employees (SE size) was included. The natural logarithm of the age of the SE (SE age) was included as a control variable, as young firms usually have higher EO than older firms (Rauch et al. 2009). In addition, controlling for the main business activity of the social enterprises. Eight business activity dummy variables were considered as control variables: (Education and Training [no = 0, yes = 1], Environment [no = 0, yes = 1], Health and Fitness [no = 0, yes = 1], Social [no = 0, yes = 1], Cultural [no = 0, yes = 1], Finance [no = 0, yes = 1], Employment [no = 0, yes = 1] and Services [no = 0, yes = 1]). The reference category in the models is Education and Training.

Results and analysis

The summary statistics of the SEs and the key respondents are reported in Table 2. The same table also incorporates a correlation matrix and provides the variance inflation factor (VIF) scores. The correlation matrix and the VIF scores suggest that there is no evidence of multicollinearity (Dormann et al. 2013). In this study, the total number of respondents was 303.

Table 1. Principal components analysis (PCA) of SEs performance.

Item	Our organization	1
Objectives	Attaining its stated goals and objectives.	0.836
Adapt to Chang	Able to adapt to the changing environment so as to attain its mission and vision during changing circumstances.	0.818
Operations	Efficient in operations	0.814
Beneficiaries	Beneficiaries satisfied (clients, donors, staff and volunteers).	0.748

Table 2. Summary statistics and correlation matrix of Performance variables (n = 303).

	Mean	SD	VIF	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Performance	5.47	1.02	–	1.00	–	–	–	–	–	–	–	–	–	–	–	–	–
2. INNO	4.60	1.47	1.66	0.33^a	1.00	–	–	–	–	–	–	–	–	–	–	–	–
3. Pro	4.30	1.22	1.90	0.31^b	0.58^a	1.00	–	–	–	–	–	–	–	–	–	–	–
4. Risk	3.78	1.50	1.38	0.20^b	0.34^a	0.48^b	1.00	–	–	–	–	–	–	–	–	–	–
5. Gender	0.58	0.49	1.23	−0.17^b	−0.05	−0.12	−0.02	1.00	–	–	–	–	–	–	–	–	–
6. Experience	0.72	0.44	1.18	0.09	0.06	0.01	0.03	0.22^b	1.00	–	–	–	–	–	–	–	–
7. HighSchool	0.07	0.25	1.18	0.03	0.04	0.02	0.01	0.00	−0.12	1.00	–	–	–	–	–	–	–
8. Diploma	0.06	0.23	1.10	0.02	0.01	−0.00	−0.07	0.05	0.06	−0.07	1.00	–	–	–	–	–	–
9. Bachelor	0.48	0.50	1.23	−0.05	−0.03	0.02	0.03	−0.02	−0.08	−0.25^a	−0.24^a	1.00	–	–		–	–
10. Master	0.28	0.45	–	0.00	0.05	0.06	0.11	−0.08	0.07	−0.16^a	−0.16^a	−0.62^b	1.00	–	–	–	–
11. PhD	0.08	0.27	1.27	0.02	−0.04	−0.10	−0.09	0.14	0.08	−0.08	−0.08	−0.30^b	−0.19^b	1.00	–	–	–
12. OtherEdu	0.02	0.12	1.05	0.07	−0.04	−0.09	−0.14	−0.05	0.02	−0.03	−0.03	−13	−0.08	−0.04	1.00	–	–
13. Log_SEs_Age	0.82	0.35	1.67	−0.16^a	−0.02	−0.01	−0.06	0.18^b	0.19^a	−0.11	0.13	0.03	−0.08	0.08	−0.01	1.00	–
14. Log_SEs_size	0.77	0.61	1.70	0.15^a	0.10	0.10	−0.05	−0.10	0.15^a	−0.09	0.03	−0.00	0.01	0.02	0.04	0.50^b	1.00
15. Educat_Tra	0.26	0.44	–	0.12	0.03	0.01	−0.02	−0.10	0.03	0.00	−0.08	0.00	−0.01	0.09	−0.02	−0.10	0.05
16. Environment	0.07	0.25	1.58	0.04	0.01	0.02	0.11	0.10	0.08	0.04	−0.07	0.05	−0.06	0.06	−0.03	0.01	0.00
17. Health_Fit	0.11	0.32	1.50	−0.03	−0.03	−0.03	−0.10	−0.17^b	−0.08	0.03	−0.06	−0.06	0.09	−0.04	−0.05	−0.10	−0.02
18. Social	0.15	0.36	1.43	−0.06	0.03	0.06	−0.10	0.08	−0.06	−0.03	−0.01	−0.01	−0.05	0.00	0.16^b	0.21^b	0.20^b
19. Cultural	0.16	0.37	1.39	0.03	−0.02	−0.03	0.04	−0.02	−0.06	−0.04	0.02	0.02	−0.02	−0.01	0.01	−0.04	0.16^b
20. Finance	−0.14	0.31	1.24	−0.14	−0.17^b	−0.11	−0.05	0.10	0.01	0.04	0.10	0.01	−0.05	−0.03	−0.04	0.14	−0.04
21. Employment	0.02	0.27	1.23	0.02	0.11	0.05	0.08	−0.02	0.01	−0.02	−0.02	−0.02	0.05	−0.05	−0.04	−0.04	−0.03
22. Services	−0.03	0.24	1.23	−0.02	0.05	0.03	0.13	0.10	0.12	−0.01	0.01	0.01	−0.08	−0.08	−0.03	−0.08	−0.06

Table 2. Continued.

15. Educat_Tra	1.00
16. Environment	−.16^b	1.00
17. Health_Fit	−0.21^b	−0.10	1.00
18. Social	−0.25^b	−0.11	−0.15^b	1.00
19. Cultural	−0.26^b	−0.12	−0.16^b	−0.19^b	1.00
20. Finance	−0.20^b	−0.09	−0.12	−0.14	−0.15^b	1.00
21. Employment	0.18^b	−0.08	−0.11	−0.12	−0.12	−0.10	1.00
22. Services	−0.15^b	−0.07	−0.09	−0.11	−0.11	−0.09	−0.07	1.00

^aSignificant at the 0.01 level. ^bSignificant at the 0.05 level.

Multiple regression analysis to test hypotheses 1, 2 and 3 (Greene 2017) was used. Model 1 in Table 3 reports results when only the control variables are included in the model. Models 2, 3 and 4 add one at a time the three independent variables to the control variables. Model 5 reports the results when all three independent variables are included together with the control variables. The Adjusted R² ranges from 0.105 to 0.204. The F tests are all statistically significant at the 0.01 level in all the models showing that taken together there is a joint relationship between the variables included in the models against performance.

Table 3. Regression models of SEs Performance.

	Model 1	Model 2	Model 3	Model 4	Model 5
Control Variables
High School	0.23 (0.25)	0.17 (0.24)	0.21 (0.24)	0.25 (0.24)	0.19 (0.23)
Diploma	0.28 (0.25)	0.26 (0.24)	0.29 (0.24)	0.35 (0.25)	0.30 (0.24)
Bachelor	0.02 (0.13)	0.04 (0.13)	0.03 (0.13)	0.04 (0.13)	0.05 (0.13)
PhD	0.17 (0.13)	0.22 (0.21)	0.28 (0.21)	0.26 (0.22)	0.29 (0.21)
Other Education	0.41 (0.45)	0.55 (0.43)	0.69 (0.43)	0.64 (0.44)	0.72 (0.43)^c
Experience	0.30 (0.14)^b	0.26 (0.13)^b	0.28 (0.13)^b	0.29 (0.13)^b	0.25 (0.13)^b
SEs Age	−0.86 (0.20)^a	−0.81 (0.19)^a	−0.83 (0.19)^a	−0.84 (0.19)^a	−0.81 (0.19)^a
SEs size	0.45 (0.11)^a	0.39 (0.11)^a	0.41 (0.11)^a	0.45 (0.11)^a	0.39 (0.11)^a
Gender	−0.26 (0.12)^b	−0.24 (0.12)^b	−0.20 (0.12)^c	−0.24 (0.12)^b	−0.21 (0.12)^c
Health_Fit	−0.29 (0.20)	−0.25 (0.19)	−0.25 (0.19)	−0.23 (0.19)	−0.22 (0.19)
Social	−0.21 (0.19)	−0.23 (0.18)	−0.27 (0.18)	−0.19 (0.18)	−0.24 (0.18)
Cultural	0.05 (0.18)	0.06 (0.17)	0.06 (0.17)	0.02 (0.18)	0.05 (0.17)
Finance	−0.34 (0.21)	−0.20 (0.20)	−0.26 (0.20)	−0.33 (0.21)	−0.21 (0.21)
Employment	−0.06 (0.23)	−0.16 (0.22)	−0.10 (0.22)	−0.12 (0.22)	−0.17 (0.22)
Services	−0.20 (0.26)	−0.24 (0.24)	−0.23 (0.25)	−0.30 (0.25)	−0.28 (0.24)
Environment	0.07 (0.24)	0.07 (0.23)	0.04 (0.23)	−0.02 (0.24)	0.02 (0.23)
EO variables
Innovation	–	0.20 (0.04)^a	–	–	0.12 (0.04)^a
Proactiveness	–	–	0.24 (0.05)^a	–	0.13 (0.06)^b
Risk taking	–	–	–	0.14 (0.04)^a	0.05 (0.04)
Constant	5.80 (.27)^a	4.87 (.26)^a	4.71 (.29)^a	5.24 (.26)^a	4.43 (.30)^a
F-value	3.21^a	4.99^a	4.94^a	3.90^a	5.06^a
ΔF	3.21^a	28.54^a	27.91^a	12.89^a	12.86^a
R^2	0.15	0.23	0.23	0.19	0.25
Adjusted R^2	0.11	0.18	0.18	0.14	0.20

^aSignificant at the 0.01 level. ^bSignificant at the 0.05 level. ^cSignificant at the 0.10 level The reference comparison variables are Education, Master Degree; and main business activity, Education and Training..

In models 2 and 5 the EO Innovation variable is strongly statistically significant at the 0.01 level and this provides evidence in support of hypothesis 1. Thus, social enterprise innovation is positively associated with SEs performance. In models 3 and 5 the EO Proactiveness variable is also statistically significant at the 0.05 level and shows support for hypothesis 2. Thus, social enterprise risk-taking is positively associated with SEs performance. In model 4 the EO Risk variable is statistically significant at the 0.01 level but in the full model 5 it is not statistically significant at the 0.05 level, or better, and thus there is no evidence to support hypothesis 3.

Several control variables are significant at the five percent level, or better, in Table 3. Experience is positively related to SEs performance at the 0.05 level. SEs’ age was negatively related to SEs’ performance at the 0.01 level. SE size was positively related to SEs performance at the 0.01 level. While the social entrepreneur gender was negatively related to SEs’ performance at the 0.1 level.

Discussion

The results have found strong support for positive relationships between EO innovation with SE performance and also between EO proactiveness with performance, but no evidence of a relationship between EO risk and performance.

Generally, SEs face resource constraints (Desa and Basu 2013) as they operate in environments where resources are scarce and expensive (Zahra, Newey, and Li 2014). Therefore, the bundling of resources through building capabilities is important for the success and sustainability of the SEs like any other firm where a combination of resources is crucial for their performance. This study suggests that SEs need to be innovative and proactive to be able to perform well, as they are key factors of their successes.

A traditional commercial enterprise will be more concerned with the end goal of economic wealth creation, whereas the SEs will be more concerned with social wealth creation (Mair and Marti 2006). It is generally noted that SEs tackle social challenges in non-traditional ways. According to Waddock and Post (1991, 393), EO at the company level improves SEs, a particular trait of theirs described as ‘private sector citizens who play critical roles to bring about “catalytic changes” in the public sector agenda and the perception of certain social issues’. The leadership characteristic used by Waddock and Post (1991) to distinguish SEs from other leaders is their capacity to outline intricate social matters to create a sense of significance that goes beyond economic performance to construct significant social performance. Social and commercial enterprises have different performance objectives; therefore, they face different challenges and different kinds of risks such as losing their credibility and legitimacy in their local communities (Hoogendoorn, Zwan, and Thurik 2011).

The results of this study have found no support for positive relationships between EO risk-taking with performance. The fact that economic goals are not the first priority of SEs combined with the fact that they usually function on limited resources explains why these enterprises are not ready to take risks by investing heavily into social projects; they would rather take a careful approach. This is supported by Weerawardena and Mort’s (2006, 29) examination of social enterprises stating that ‘[t]he majority of the cases appear to adopt a highly cautious approach in dealing with risk having a clear focus on survival of the organization’. This idea that SEs are less risk tolerant than CEs has also been supported by Abu-Saifan (2012), who drew upon a large number of scholarly studies to differentiate CEs from SEs. The author suggested that though both categories are risk-takers, CEs tend to be risk bearers, and SEs tend to be highly accountable. As mentioned by Dess (1998) about a decade before, Abu-Saifan (2012) also implied that SEs work boldly with limited resources; however, there is a severe lack of risk-taking in social enterprises. In other words, SEs often do not have sufficient funds to grow because they need to invest in new ventures to grow, but because of a lack of funds, they cannot risk the funds that they already have. Therefore, they are not able to invest in new ventures, and so they are not able to grow (Emerson et al. 2007). The limited sources of funding available for SEs in comparison with CEs create greater uncertainty association to funding which push SEs to assess the risks involved to sustain their firms (Weerawardena and Mort 2006).

This study agrees with Anderson et al. (2015) and their conceptualisation of EO by collapsing the innovativeness dimension with the proactiveness into one dimension of ‘entrepreneurial behaviors’ and this research suggests that it is an appropriate conceptualisation to be used in SE literature and as a promoter of SEs performance. Entrepreneurial behaviors will aid SEs to gain the support as well as acceptance of multiple stakeholders while seeking to achieve the enterprise’s social objective to create the necessary social impact and represent potential success factor and outstanding performance.

Generalisation of the results

The validation analyses conducted here indicate that the construct was, in fact, valid and that the findings would also be beneficial to other areas of SEs as well. The application scope of these findings can be extended to other countries. First, Saudi Arabia is part of the Middle East and North Africa (MENA) countries where the socio-economic landscape is similar, therefore, people in those countries face similar challenges such as high unemployment rates and gender inequality (Jamali and Lanteri 2016). Secondly, Saudi Arabia is considered a developing country and Saudi SEs face similar resources constraints as other SEs in other developing countries. Researchers have found that even SEs in developed countries face challenges in resources mobilisation (Austin, Stevenson, and Wei‐Skillern 2006; Desa and Basu, 2013). In this context, this study advances existing knowledge in the context of EO and SEs’ performance in developing countries, setting theoretical and empirical foundation for a better understanding of such phenomena in developed countries as well.

Although EO is one of the few entrepreneurship constructs applied differently across countries and regions because of the differences in business cultures, it cannot be ruled out as totally dependent on this factor. This paper has argued that the inconsistencies in findings concerning the association between the EO dimension of risk-taking and SEs’ performance are due to the differences between the entrepreneurship and the SE contexts and have found empirical support for this argument. Therefore, it is important to interpret the results of this study in the context of Saudi Arabian SE endeavours and, when applicable, to implement them in other contexts.

Further research is needed to establish to what extent such findings could be applied to contexts other than specific SE in contemporary Saudi Arabia. To examine if the results can be expanded to other business contexts, there are some obvious limitations to the study, and the focus of the next section is particularly on those limitations that restrict the generalizability of the results.

Limitations and future research

Like any other study, this research has limitations that open avenues for further inquiry. The study does not enable us to determine causal relationships in the strict meaning of the concept because both the dependent and independent variables were examined at a specific point in time, rather than a wider span of sequential or arbitrary moments chosen in advance. This approach limits the possibility of inferring full causality from emerging relationships; therefore, a longitudinal study would open a new avenue of research that would enrich the findings of the present study.

In highlighting the unique ways in which EO may apply to the particular SE contexts, it is suggested that the obvious differences in EO dimensions require unique conceptualisation and that the measurements require unique instruments in order to appropriately account for these differences. Therefore, a modification of the EO scale to be more consistent with the context of SE is suggested. For example, risk-taking items need to be changed to fit this specific context considering that SEs may not take such large risks as do commercial entrepreneurs because with increased risk there is an increased potential for failure. This means two things: first, SEs could lose their legitimacy, and second, if SEs go bankrupt, the beneficiaries of their services may not be helped by others in the market (e.g. dying people, poor people receiving health care, those receiving SE-provided education).

This research suggests that research in the same country will take a sample of CEs and SEs and compare their EO against each other and find out on which dimensions each sample will have higher levels. After doing so, the researchers may compare gender differences in the sample and compare their performance too. Such research would firstly help us understand more about the similarities and differences between CE and SE; secondly, it would help us understand gender differences that will help policy makers to create a better ecosystem for fostering both CE and SE. In the future, larger studies with longitudinal statistical evidence would be helpful in this particular field because these issues need to be resolved to advance a more accurate assessment of EO and SE.

Conclusion

This paper has presented new insights into an under-researched area, the links between EO and SEs’ performance. A new data set of 303 SEs in Saudi Arabia have been used and the results found that innovativeness and also proactiveness but not risk-taking are related to SEs’ performance. This study would like to underline two major implications for managers of SEs. The first is that although the effects of EO vary across SEs, there is a generally positive effect of EO across the contexts investigated. In other words, based on the outcomes of this study, it appears that EO generally contributes to social enterprises’ performance. The second is that EO entrepreneurial behaviors can be used as a mechanism to overcome constraints imposed by limited resources and an environment in which new opportunities rarely occur. It is under these conditions that managers can truly benefit from being innovative and proactive, thus gaining legitimacy and expanding their business’ respective stockholder pools.

Disclosure statement

No potential conflict of interest was reported by the authors.