Introduction

The research on social enterprise (SE) has predominantly focused on defining what it is (Bacq and Janssen 2011; Choi and Majumdar 2014), the associated organizational processes and management (Desa and Basu 2013; Katre and Salipante 2012), measurements (Lepoutre et al. 2013; Stevens et al. 2015), politics (Mason 2012; Dey and Teasdale 2013), and institutional environments (Kerlin 2013; Littlewood and Holt 2015) vis-à-vis non-profit and for-profit organizations. In most of the literature, SE tends to be conceptualized in the form of hybrid Footnote 1 organizations because of the ways in which it combines multiple institutional (i.e., social welfare and commercial) logics (Cooney 2006; Doherty et al. 2014).

The literature reveals that creating hybrid organizations is challenging because founders will encounter tension when balancing the contradictory social welfare and commercial logics. This tension may manifest itself in a “sense of dissonance” (Stark 2011, p. 14) for the founders of social enterprises, possibly leading to an identity crisis and even “mission drifting” (i.e., turning into for-profit venture, Ebrahim et al. 2014, p. 84). As such, only certain individuals are seemingly willing and/or capable of combining both logics. Scholars have argued that a connection potentially exists between the founders’ identities (i.e., their roles and personal identities) and the ways in which they combine multiple institutional logics (Wry and York 2015). Similarly, other scholars underscore the importance of understanding founders’ past life experiences in shaping their likelihood of creating social enterprises (Christopoulos and Vogl 2015; Germak and Robinson 2014; Yiu et al. 2014). These studies suggest the need to deepen our understanding of the individuals who establish social enterprises as hybrid organizations as well as their life story and motivations.

There are two major assumptions about the individuals who establish SEs as hybrid organizations. One assumption envisages social entrepreneurs as heroic agents, who have the power to enact social change, and emphasizes their success stories (or the normative construction of SE, Bacq et al. 2016; Nicholls 2010) on the one hand; and the spontaneous, grassroots-, and community-based SE that emphasizes the power of the collectives in enacting social change (or the positive construction of SE), on the other hand. Another assumption is the distinction between push and pull factors (e.g., experiencing a social problem versus career development; Yitshaki and Kropp 2016; Yiu et al. 2014) in SE emergence. These assumptions, while being important, tend to dichotomize the construction and driving forces of SE emergence and ignore the complexity of SE as hybrid organizations.

Despite the popularity of hybrid organizing as a major theoretical perspective of SE (Cooney 2006; Dart 2004; Doherty et al. 2014), little research has investigated the biographicalFootnote 2 antecedents of the emergence of SE in the form of hybrid organizations or that which enables the integration of multiple institutional logics. Moreover, extant research on SE emergence tends to lack larger qualitative studies (e.g., Germak and Robinson 2014; Perrini et al. 2010) to enable deeper generalization of the findings. In other words, we know little about the driving forces or factors that enable individuals to combine contradictory institutional logics within an organization. What combination of factors will lead an individual down a path toward SE? Will a combination of certain factors be more important than others? For instance, will a young social activist need an MBA and further qualifications and resources to start a social enterprise? Will a business executive need volunteering and additional social activism experience before starting a social enterprise? Research on this gap will allow us to unpack the factors that explain and predict individuals’ tendencies to start SEs and engage in SE and to advance hybrid organizations as its theoretical foundation. This research may also offer valuable insights into policymakers and funders who select and support social entrepreneurs.

In this article, we ask an important research question: What are the biographical antecedents of the emergence of social enterprises? To answer this question, we employ the emerging narrative approach in SE research (Chandra 2016a; Froggett and Chamberlayne 2004; Teasdale 2012) and entrepreneurial narratives research more broadly (Lieblich et al. 1998; Martens et al. 2007) to unpack micro-, meso-, and/or macro-level factors that may explain individuals’ tendencies to create social enterprises in the form of hybrid organizations. Narratives about the emergence of SE often contain rich information that may reveal interesting stories and new dimensions to explain the whys and hows of hybrid organization creation. To examine narratives regarding the emergence of SE, we inductively study the biographical profiles of self-identified social entrepreneurs who have been supported by two of the world’s largest SE support organizations: Ashoka and the Schwab Foundation. Prior to data collection and analysis, we familiarized ourselves with the SE literature, particularly the factors that drive SE, and the factors that drive commercial and social venturing activities as a theoretical backdrop. Using the Gioia’s methodology, we inductively code the biographical profiles of Ashoka and Schwab social entrepreneurs (n = 200) and subsequently classify them using principal component analysis (n = 117), thereby contributing to the SE literature in two important ways.

First, we found eight biographical antecedents of the emergence of social enterprises as hybrid organizations (in descending order of importance for each component): collectivism, ideologism, altruism, and spiritualism, which can be aggregated as the social skills component; and entrepreneurialism, resources, professionalism, and higher education, which can be aggregated as the economic skills component. The “economic skills” is a proxy for what neo-institutional theorists call “social position,” which provides the willingness, resources, and legitimacy to enact social change. This finding offers finer explanations that go beyond the conventional push or pull theorization in SE and the motivation for SE. Second, our study is the first to demonstrate that social entrepreneurs are not a homogeneous group but are highly heterogeneous in their mix of social skills and social position. We develop a typology of SE-as-hybrid-organization: the social elite, the social-grassroots, the elite, and the commoner and offer illustrations using biographical narratives. Our typology suggests that there is no “ideal” or “typical” type of social entrepreneurs and that SE-as-hybrid organization has a pluralistic origin which comprises individuals with different social skills and social position who start SE and enact social change. We finally present the conclusion, limitations, and avenues for future research.

Literature Review

Defining Social Enterprise

Although no single definition describes what a social enterprise is (Choi and Majumdar 2014), a growing consensus argues that it is a hybrid organization that combines commercial and social welfare logics (Cooney 2006; Dart 2004; Doherty et al. 2014). Logic here refers to the accepted goals of a particular sector and the means by which they are pursued (MacDuffie 1995). On the one hand, the usual dichotomy is that businesses seek to optimize profit by engaging in trade, while charitable foundations or philanthropic organizations address social problems through aid or donations. On the other hand, social enterprises combine these two seemingly opposed goals and means—they attempt to achieve their social or environmental missions by engaging in commercial activities, thereby representing a hybrid form. As hybrids, they constitute what organizational scholars call institutional innovations (Cajaiba-Santana 2014).

Who are the individuals behind these social enterprises? Social entrepreneurs have been described in rather lofty terms, albeit differently in various fields. In the public administration realm, social entrepreneurs are described as private sector citizens who seek to make “catalytic changes” in the public sector agenda (Waddock and Post 1991, p. 393) and those who explicitly seek to address market failures across many sectors and to solve other societal problems, including problems that concern the government (Terjesen et al. 2015; Bielefeld 2015). Management scholars, such as Zahra et al. (2009), describe social entrepreneurs as those who “make significant and diverse contributions to their communities and societies, adopting business models to offer creative solutions to complex and persistent social problems” (p. 519). Rather than describing social entrepreneurs as a bundle of traits or characteristics, Dacin et al. (2011) propose that “the primary mission of the social entrepreneur being one of creating social value by providing solutions to social problems—holds the most promise for the field” (p. 1204) and argue that social value creation is often closely linked to economic outcomes that, in turn, produce financial resources to accomplish a social mission (Dacin et al. 2010). Santos (2012) argues that maximizing both value creation and value capture in an organization is difficult and that organizations must have a “predominant focus” (p. 338). He calls for a better definition, as social entrepreneurs are individuals who seek to optimize value creation and achieve value capture. Similarly, the two major SE support foundations, Ashoka and Schwab, characterize them as innovators who find new solutions to society’s most pressing problems.

The Narrative Construction of Social Entrepreneurs: Heroic and Community

One avenue for advancing SE research, particularly with regard to the why and how individuals are willing and/or able to embrace multiple institutional logics, involves studying the narratives that explain the emergence of SE in the form of hybrid organizations. Narratives about why and how social enterprises are created can help unpack important forces and factors that explain why and how certain individuals are willing and able to combine contradictory institutional logics within an organization. This perspective of SE resonates with the “entrepreneurial narrative” approach (Lieblich et al. 1998; Martens et al. 2007) and particularly represents an opportunity to extend the emerging “narrative approach” in SE research (Chandra 2016a; Ruebottom 2013; Yitshaki and Kropp 2016). Studying the narratives regarding the emergence of SE also reveals social entrepreneurs’ life story and motivations to engage in SE (Froggett and Chamberlayne 2004).

Extant narrative SE research comprises the construction of social entrepreneurs as heroic- and community-driven agents of change (Bacq et al. 2016; Nicholls 2010). In the normative construction of SE, social entrepreneurs are often portrayed as heroic individuals or lone agentic actors with certain power, capabilities, and resources (Campbell 2004) as well as their success stories as being central to SE. Examples include stories of leaders of Drug Free America and Hands Across America who created catalytic change by addressing poverty and drug abuse, respectively (Waddock and Post 1991) and how Hong Kong-based Diamond Cab SE creatively developed Asia’s first specialty taxi market for wheelchair users (Chandra 2016b). Ashoka and Schwab are among the well-known foundations and fellowship organizations that help promote the heroic depiction of social entrepreneurs of the change makers (Nicholls 2010). Nicholls (2010) argued the focus on heroic and successful stories of SE in SE support organizations like Ashoka was influenced by the venture philanthropy model (“commercial logic”) employed by such organizations.

The construction of social entrepreneurs as community-driven change makers has also gained traction in recent years (see Maclean et al. 2013; Montgomery et al. 2012). From this lens, SE is seen as a highly collaborative process and the role of community that enables the pooling and trading of resources. Therefore, the community and other social actors—from public, private, third sectors and social movements—tend to get more attention and credit in explaining the emergence and success of SEs. This includes how the founders and management team of Italy-based San Patrignano, a drug rehabilitation community SE, relied on supportive networks and community members as resources in starting and growing the SE (Perrini et al. 2010) and how US foundations developed collaborative solutions to improve population health (Heinze et al. 2016). Community-driven SE tends to be conceptualized broader than the conventional SE to include various forms of collective action, from social movements (e.g., wind energy and grass-fed meats movements), rural cooperatives, and cross-sectoral partnerships that serve social purposes (e.g., Habitat for Humanity and Kiva; Montgomery et al. 2012).

Overall, SE scholars tend to dichotomize the narrative construction of SE as either heroic- or community-focused. This stands in stark contrast with the SE and entrepreneurship literature that recognize the role of the (creative, resourceful) individuals (Campbell 2004; Waddock and Post 1991) and the collective process (Birley 1986; Montgomery et al. 2012; Nicholls 2010) in organizing and creating social change. To date, little research examines the factors that explain the emergence of SE on a large scale and the promise of the narrative approach in explaining how social entrepreneurs are willing and/or able to integrate different institutional logics.

Antecedents of Social Enterprise Emergence

Prior SE research highlights the social welfare logic part of SE by underscoring the role of compassion, altruism, and other-oriented drivers in the creation of social enterprises. Miller et al. (2012), for instance, theorized a link between compassion and SE, arguing that compassion works as a prosocial motivator that triggers cognitive mechanisms that increase the likelihood of individuals engaging in SE. Yiu et al. (2014) tested the altruism-SE nexus using China’s SE database, wherein they reported that past distressing experiences push individuals to participate in the sector. Other studies (e.g., Christopoulos and Vogl 2015; Germak and Robinson 2014) have similarly reported evidence of social welfare logic, including a desire to help society, a focus on nonmonetary benefits, and exposure to social problems, as the drivers of SE.

By contrast, Lee and Battilana’s study (2013) highlights the commercial logic part of SE. They proposed and tested a theory that social enterprise emergence is strongly influenced by an individual’s past work experience. They found evidence that, when an individual has had professional exposure to a commercial enterprise, albeit not for an overly long period, they are more likely to create hybrid social ventures (i.e., social enterprises) as opposed to ones that only employ the social welfare logic, such as non-profit organizations.

The emerging research on SE narratives may provide further insights into the emergence of SE. For instance, Chandra’s (2016a) study of the narratives of social and commercial entrepreneurs revealed that social entrepreneurs used more words associated with other, stakeholder and justification orientations and fewer words associated with self-orientation. However, these two types of entrepreneurs did not differ in their use of words associated with an economic orientation. Ruebottom (2013) studied the rhetorical strategies that social entrepreneurs use to overcome institutional barriers and found that protagonist versus antagonist meta-narratives were well integrated to create tension and to persuade stakeholders of the social enterprises’ legitimacy. Parkinson and Howorth’s (2008) study of the discourse surrounding “doing” SE revealed that social entrepreneurs are preoccupied with local issues, collective action, geographical community, and local power struggles; this study also showed how the tensions between social morality and managerial rationality may surface in social entrepreneurs’ language. These findings suggest the presence of social welfare logic (i.e., other and stakeholder orientation, local issues, collective action), commercial logic (i.e., economic orientation), and political logic (e.g., local power struggles) as parts of SE narrative.

Recently, scholars argue for the push versus pull factors (Yitshaki and Kropp 2016), as an alternative theorization of SE emergence. In fact, all the factors discussed above can be categorized into push and pull factors. For example, push factors include seeing and experiencing social problems (Yitshaki and Kropp 2016; Yiu et al. 2014) and feeling dissatisfied with the status quo (Chandra 2016b); and pull factors include pursuing career development opportunities in SE after many years of work experience in the non-profit sector or being attracted to SE due to the influence of friends and families.

The important commonality in the literature cited above is the argument that the emergence of SE is driven by different processes and factors than those of commercial entrepreneurs, on the one hand, and social workers, activists, volunteers or non-profit founders, on the other hand, whose orientations are associated with either profit or social welfare, respectively. The paths toward SE may be similar to and different from those of other types of ventures, but much remains unknown. Therefore, an in-depth exploration of the antecedents of the emergence of SE and the use of a relatively large sample size is necessary to increase our knowledge of SE emergence and to push SE scholarship forward. We explain our methodology in the next section.

Research Methodology

To answer the research question, we first employed Gioia’s methodology (Gioia et al. 2013), which is appropriate for inductive theory building and the narrative approach (Dempsey and Sanders 2010; Froggett and Chamberlayne 2004; Lieblich et al. 1998; Pentland 1999). Although existing theory informs our research, we follow Pratt (2009) by distancing ourselves from such theory to generate new insights. Subsequently, we conducted content analysis (Krippendorff 2004) and performed principal component analysis (Lê et al. 2008; Sebastien et al. 2008) on the coded narratives.

Narratives contain sequences or patterns of events, focal actors’ roles, social networks and demographics, a voice or point of view, a moral context (e.g., beliefs and values), and other contexts (Pentland 1999). Narratives are a form of “process data,” and they have been described as highly accurate, though not overly simple or general (Langley 1999). Narratives allow us to study a phenomenon from a relational ontology, that is, actions and interactions as they emerge (Garud and Giuliani 2013). Narratives provide comprehensive information about how actors construct their lives and identities by the references they made to past, present, and future actions (Lieblich et al. 1998). Narratives provide powerful tools to explore what entrepreneurs (or others) say about what they do (Gartner 2007); reflect the subjective sense making of actors’ action and emotion (Weick et al. 2005) as well as remembered facts and a presentation of people’s selves rather than accurate scientific truth (Yitshaki and Kropp 2016).

Our narrative data comprise the biographical profiles of social entrepreneurs—or so-called fellows—who are supported by the Ashoka and Schwab Foundations, two leading support organizations for SE that provide mentoring, funding, and networking. Much can be learned from a biographical profile, as it reflects the life history of an individual at the nexus of social context, personal character, and past experiences (Davies et al. 2002; Lieblich et al. 1998; Smith 1994) that cut across micro-, meso- and macro-levels, all of which can shape decisions, actions and subsequent choices. Biographical profiles contain more than just the traits (Gartner 1988) of entrepreneurs but also the key events in their life, people involved in a venturing process, resources they draw from, the context where an entrepreneur operates, etc., which enables a holistic view of the emergence of SE. The biographical profiles of Ashoka and Schwab fellows were written from the social entrepreneur’s perspective, based on multiple stages of selection process (to be described in more detail in the “Sample” section) and information gathering via interactions between the SE support organizations and social entrepreneurs, but they were described using a consistent format and were maintained by the SE support organizations.

Sample

The advantage of using Ashoka and Schwab fellows’ biographies as data sources is that they have a high level of credibility and quality. That is, these profiles capture self-identified social entrepreneurs who have gone through a rigorous selection process to become “fellows.” For Ashoka, this comprises a five-stage selection process involving nomination, first opinion and second opinion interviews and site visits, panelists’ review and finally the board review to assess the founders’ commitment and match with the Ashoka’s espoused values and criteria (see https://www.ashoka.org/en/engage/recommend/fellow). Schwab uses a three-stage selection process comprises dossier screening by the Foundation and local experts review, followed by interviews and the selection committee review to select winners (see http://www.schwabfound.org/content/selection-process).

Ashoka has had nearly 3000 fellows, while the Schwab Foundation for SE has had over 300 fellows to date. At the time of the data collection (July 2014), the profiles of 2410 Ashoka fellows and 312 Schwab fellows were extracted from their Web sites (www.ashoka.org; www.schwabfound.org/entrepreneurs). Two sample biographical texts of Ashoka and Schwab fellows are shown in Web Appendix 1. There is a high degree of consistency exists between the two sources in terms of the format of the biographical narratives. These narratives provide information on the problems that fellows seek to solve; the new ideas, solutions, innovations, and activities that they present to solve these problems; and their backgrounds. As shown later in this article, our inductive coding process considers not only the “person’s background” but also any relevant information in the biographical texts.

From this pool of raw data, we randomly selected the biographical texts from Ashoka (n = 100) and Schwab (n = 100) and used a stratified sampling method to obtain a high degree of representativeness (Neuman 2005). We summarize the samples in Table 1. Ashoka fellows from Asia, South America, and North America constituted the top three of the five regions of Ashoka fellows’ countries of origin, and the proportion of fellows was balanced across the six fields of work (i.e., civic engagement, economic development, environment, healthcare, human rights, and learning/education). We thus sampled them accordingly (see Table 1). In their selection and display of particular fellows on their respective Web sites, Ashoka and Schwab label and categorize these fields of work to pre-classify the fellows into different practice areas.

Table 1 Sampling for qualitative data analysis
Full size table

Schwab categorizes its fellows into 21 fields of work and eight different regions, and we sampled them proportionally (see Table 1). Asia, cross-regions, and South America remain the three largest regions sampled, and Table 1 presents the distribution of the Schwab fellows across these 21 fields of work.

Qualitative Data Analysis

The randomly selected Ashoka and Schwab biographical texts extracted from the Web sites were saved in plain text (.txt) files. The biographical text files were then imported into a computer-assisted qualitative data analysis (CAQDAS) software. CAQDAS-based qualitative research can enhance the transparency and trustworthiness of qualitative research and allow researchers to study a large sample size and to share the coding outputs among research collaborators and journal reviewers/editors (Sinkovics and Alfoldi 2012). For this study, we use RQDA (see Chandra and Shang 2017; Huang 2015), a sophisticated CAQDAS tool that runs on R programming language, an extension of the open-source R Project for Statistical Computing application (R Development Core Team 2009).

Prior to the actual analysis, two of the research team members independently coded the biographical texts by randomly selecting 50 biographical profiles/texts on Ashoka and Schwab fellows from their respective Web sites and manually reading each of them. This process was important to help us familiarize ourselves with the nature of these biographical texts.

Next, we imported these text files into RQDA software, and two of the research team members independently conducted “open coding” (Corbin and Strauss 1990; Gioia et al. 2013) by first coding the Ashoka biographical profiles and then coding the Schwab profiles. To avoid information loss during the coding process, we sought to code in a way that would preserve the original meaning of the first-order codes and only aggregate them in the theoretical realm in the next orders of analysis. We also compared the codes for the Ashoka profiles with those for the Schwab profiles, although our interest and analysis were focused on the aggregate samples, which allowed for a large-scale analysis and better external validity. This process was iterative and extensive and generated 1316 first-orderFootnote 3 codes. We cross-validated the consistency and accuracy of the fellows’ narratives with other publicly available online sources (i.e., personal Web sites, social enterprises’ official Web sites, news articles, and magazines) by manually reading and cross-referencing to ensure the convergence of facts about events, actors, relationships, and opinions across all the fellows in the sample. We found that the biographical information from the Ashoka and Schwab Web sites and other online sources was factually consistent for all fellows.

The research team further amended the codes, and after 10 meetings over a 1-year period, the team reached a consensus with regard to the themes emerging from the data, then combined all first-level codes into eight second-order categories, and finally reduced them to two aggregate dimensions. In doing so, we cycled between the codes and the literature (Gioia et al. 2013) to make sense of the findings and to better ground them in existing research. Throughout this process, another research team member acted as the “devil’s advocate” to challenge and question the interpretations of the data, as suggested by Gioia et al. (2013). Therefore, inter-team discussions were a key part of the sense making process in understanding the findings and positioning the findings in the study. We also presented and discussed the findings with five scholars from different fields—in SE, sustainability, entrepreneurship, nonprofits and public administration—to obtain a more well-rounded interpretation of the findings.

Qualitative Findings

We found eight antecedents (i.e., altruism, spiritualism, collectivism, ideologism, entrepreneurialism, professionalism, resources, and higher education) of SE emergence which can be categorized into two aggregate dimensions: social skills and economic skills. Social skills dimension comprises altruism, spiritualism, collectivism, and ideologism. Economic skills dimension comprises entrepreneurialism, professionalism, resources and higher education. We depicted a data structure that demonstrates the process in which we started with the first- to second-order code categories and then established the final aggregate theoretical dimensions in Fig. 1. Only representative “quotations” are presented in the first-order codes for parsimonious reasons. We discuss each of the SE antecedents in sequential order below (as in Fig. 1). Additional representative data (c.f. Fig. 1) is shown in Web Appendix 2.

Fig. 1
figure 1

Data structure from qualitative data analysis

Full size image

Social Skills

Social skills are the competencies, values and resources that a social entrepreneur acquired through social interactions which sensitize him or her toward the plight of others or oppressed groups of people in the community. We discuss each of the four sub-dimensions of the social skills below.

Altruism

The altruism concept appears prominently in our qualitative data analysis. Our analysis suggests that altruism may come from various sources, such as (1) family influence, (2) distressing personal experience, (3) contacts with the disadvantaged and (4) volunteering experience, which extend existing knowledge on the SE motivation (Germak and Robinson 2014; Miller et al. 2012).

This study demonstrates that family influences (particularly that of parents and close relatives) are important in inspiring people to engage in SE. Our data show that the formation of a social entrepreneur’s altruistic values may be influenced by his or her family members’ values. As such, social entrepreneurs may absorb their family’s socially oriented values, which may directly or indirectly inspire them to engage in SE. One representative example is Suprabha Seshan (Ashoka India, Environment; www.ashoka.org/en/fellow/suprabha-seshan), who developed a deep sense of belonging in the natural world and later started Gurukula Botanical Sanctuary that conserves the environment and conducts ecological advocacy because of the influence of her parents’ interests in the environment and the suffering of refugees.

A distressing personal experience can also act as a triggering mechanism for SE. Our data show that many social entrepreneurs experienced a tragedy, misfortune or disaster in his or her personal life, which eventually triggers a sense of altruism for others. For example, Socorro Guterres (Ashoka Brazil, Human Rights; www.ashoka.org/en/fellow/socorro-guterres), who started the Black Culture Maranhão SE to end racial discrimination against the black population in Brazil, was born in a poor family and experienced racial discrimination from an early age and her own daughter was a victim of racial slur and physical violence in the school because of her skin color.

Contact with the disadvantaged is also another important factor that triggers the path toward SE. Our data show that many social entrepreneurs had contacts with poor or deprived or oppressed people in their early life, which led to a sense of empathy and altruism to help alleviate the sufferings of the disadvantaged. One example is Ann Cotton (Schwab UK, Education/Children and Youth/Women; www.schwabfound.org/content/ann-cotton) whose contact with girls in rural Zimbabwe during her research trip changed her view about life and this led to a life-long quest to help poor girls receive education. As stated in her Schwab profile:

Ann Cotton was first inspired to change the future of girls in rural Africa during a research trip in a remote village in Zimbabwe in 1991. What she discovered there – that girls’ exclusion from education [due to poverty] was culturally based – profoundly changed her view. Moved by this experience, she founded Camfed and has worked ever since to ensure that poor girls are given the chance and resources to go to school. (emphasis added)

Another important driver of social entrepreneurs’ altruism is volunteering experience. Many of the social entrepreneurs in our data had volunteering experience, which triggered their sense of altruism to help the disadvantaged. Importantly, volunteering can also be influenced by family values (i.e., parents’ altruism spreads to children). For example, Michelle Lem (Ashoka Canada, Healthcare; www.ashoka.org/en/fellow/michelle-lem), who started a SE that provides animal welfare services, had extensive volunteering work experience for disabled and homeless people and sick animals; this was encouraged by her immigrant parents who were medical practitioners, as illustrated in the Ashoka:

Her [Michelle’s] mother was a nurse and her father, a dentist and well known community activist…in Canadian history. Both of her parents encouraged volunteerism at a very early age, and Michelle would spend a substantial amount of her volunteer time working with people with disabilities, then later in her career at homeless shelters… After graduation [psychology, microbiology and veterinary medicine]…, Michelle started several small businesses including an online pet sympathy card business to help. veterinarians express empathy after the loss of a beloved pet…a companion animal mobile unit …fill a gap in care…by establishing Community Veterinary Outreach. (emphasis added)

Spiritualism

Our study also found that spiritualism—which consist of (1) religious beliefs and (2) personal contact with a religious group—can encourage SE. Our study also shows that people with religious beliefs are very likely to have done community work, which may increase their likelihood of pursuing SE. A representative case of the religious influence concept can be seen in Marcelina Bautista (Ashoka Mexico, Human Rights; www.ashoka.org/en/fellow/marcelina-bautista), who worked as a domestic employee and experienced discrimination and exploitation. The Catholic Church and the Catholic youth movement appear to be influential drivers of her involvement in SE. As stated in her Ashoka biography:

In Mexico City Marcelina became involved with the local Catholic Church and the global Christian Youth Workers movement. At 17, she…became involved in a group of workers organized by the church, through which she voiced her frustrations and overcame her shyness. Marcelina also began to learn about human rights and workers’ rights in conjunction with her ongoing Bible study. In January 1988, she and a group of fellow domestic workers founded the group La Esperanza, (“The Hope”) with the goal of educating female domestic workers about their rights. (emphasis added).

Contact with religious groups including family members who are religious leaders also plays a role, often making individuals more aware of religious conflicts. Our study shows that the identification of religion-related conflicts prompted people to find solutions and to promote social change via SE. For example, Nerlian Gogali (Ashoka Indonesia, Civic Engagement; www.ashoka.org/en/fellow/nerlian-gogali), whose father is a Christian pastor:

[Lian’s] father taught religion…and tolerance of other religions…When she was in high school…she interviewed members of rival motorcycle gangs from different religious backgrounds. She realized the conflict always came down to the women from different religious backgrounds… While studying [theology], Lian joined DIAN Interfidei, an interfaith organization in Yogyakarta…[conducted] research…and live in the refugee camp for one year to have direct experience with Christian and Muslim women and children who had survived from the conflict… She uncovered stories from women who had helped each other regardless of religion. These stories have a great impact on her work today. (emphasis added).

Collectivism

We found that collectivism (Heinze et al. 2016; Montgomery et al. 2012) is also an important concept driving the emergence of SE. In our study, individuals’ collectivism comprises collective values and the willingness to collaborate with other groups to improve society. An example of the collectivism concept is found in Abla Al Alfy (Ashoka Egypt, Healthcare; www.ashoka.org/en/fellow/abla-al-alfy), who had done community work early on in her childhood which influenced her to improve the medical profession in her country by setting up nutrition counseling centers and nutrition counselor certification. As stated in her Ashoka biography:

During her adolescent years, Abla was involved in her community and from 1967 to 1969 she supported families who experienced loss during a recent war by collecting money and clothes and distributing them to needy families. Her passion was working with children. After [medical school and pediatrics study]… each hospital that Abla worked at she followed a personal mandate by asking herself how she could help improve the hospital, help her colleagues become better doctors, and how she could learn to serve her patients more effectively. (emphasis added)

Ideologism

Ideologism, or conformity to an ideology (Den Hond and De Bakker 2007), is another important driver of the emergence of SE. Our data show that, in many cases, SE is an extension of people’s prior (socio-political) ideology work. The findings reveal that socio-political activism activities more frequently expose individuals to social problems and, in turn, enable them to identify social gaps and problems and find solutions to the problems. An example is Sanjit Roy (Schwab India, Education/Energy/Women; www.schwabfound.org/content/sanjit-bunker-roy), a former social activist turned social entrepreneur, who embraced Gandhi’s ideology on self-sufficiency. This is shown in his Schwab profile below:

For over 40 years Sanjit Roy has demonstrated the power and impact of the grassroots community movement, and the need for social entrepreneurs to be social activists first. He was influenced by Mahatma Gandhi’s spirit of service and thoughts on sustainability….Created in 1972, [Roy’s] Barefoot College in Tilonia, Rajasthan, was inspired by the principles of Gandhi [on self-reliance] and around the concept of the village as a self-reliant unit. (emphasis added)

Another example of ideologism is Jacek Strzemieczny (Ashoka Poland, Education; www.ashoka.org/en/fellow/jacek-strzemieczny), who was driven by democracy and citizenship values, as shown in his Ashoka profile below:

[Jacek] was involved in a number of underground activities, including the creation of several independent educational institutions during the communist era in Poland… [through these activities] he concluded that the best way to achieve his overall goal of reforming Polish schools could best be achieved from outside the government…[and later he] is dedicated to promoting effective civic education for informed and responsible citizenship in new and established democracies.” (emphasis added).

Entrepreneurialism

As for SEs’ economic skills, we found that entrepreneurialism is an important concept that enables individuals to start up social enterprises. In this study, entrepreneurialism consists of an individual’s prior business experience and general leadership experience, which are two essential elements for starting up new organizations. Our study finds that business and not-for-profit experience enables individuals to combine their business acumen with social objectives and pursue their interest in social causes using an SE approach. Our data show many leaders and workers of (not-for-profit) community development organizations, former civil servants, and social workers who extended their community and public service work by establishing SEs (e.g., Aicha Ech Channa, a former Moroccan social worker, www.ashoka.org/en/fellow/aicha-ech-channa); as well as those who made the transition from business to SE. An example of entrepreneurialism is the case of Jack Sim (Ashoka Singapore, Healthcare; www.ashoka.org/en/fellow/jack-sim), who had started successful businesses, which gave him the skills, resources and legitimacy before starting his World Toilet Organization (WTO) SE. As stated in his Ashoka biography:

Their businesses include construction materials, real estate development, and an international school, among others…Jack believes that through these experiences he learned to work out partnerships, build trust, and take an interest in others’ successes. He enjoyed the human side of the business, meeting with customers and working with his staff. At his company he learned that the key to success is to build trusting relationships with everyone. Those skills form the basis of his network approach to WTO [social enterprise].” (emphasis added)

Professionalism

In this study, professionalism, or individuals’ professional experience and knowledge in either for-profit, non-profit and government (public service) organizations, trigger individuals to found social enterprises. Our data show that professional (i.e., technical, administrative) experience in these organizations that makes hybridization possible—without the professional experience in organizations, setting up a SE is difficult. For example, Adair Meira (Schwab Brazil, Children and Youth/Environment/Rural Development; www.schwabfound.org/content/adair-meira), who started SE to serve youths, had over 15 years of experience as a communication professional. He also has a long track record as environmentalist and in social campaigns. As stated in her Schwab biography:

Adair Meira is a businessman and environmentalist who has long been committed to social causes in poor areas of large cities. As a communications professional he was involved in social mobilization campaigns between 1980 and 1990, especially in struggles for children’s and adolescents’ rights. In 1994 he founded Fundação Pró-Cerrado, which has become the largest educational and income-generating youth programme in Brazil. With his experience…he created a platform that connects the needs of today’s youth with the business world, increasing awareness of the need to support young people in Brazil. (emphasis added)

Resources

Resources refer to financial resources and social capital, which in turn brings tangible (e.g., people, partners) and intangible (e.g., legitimacy) resources. We found that SEs’ resources could either come from family support or own savings and external funding, as illustrated by the case of Paul Born (Ashoka Canada, Economic Development; www.ashoka.org/en/fellow/paul-born), who was born in an economically well-off (i.e., elite) family and had substantial personal resources to start an SE. As stated in his biography:

At the age of 6, Paul’s family quickly moved from being very poor to being economically well-off due to the increased value of their land holdings and successful farming ventures…. At the age of 16, Paul started his own chicken hatching business…his business to have more than 82 employees…moving more than 10,000 chickens in a day… [At 19] Paul had already sold his business…to study religion…. Leaving the seminary two years later, Paul…served his community with Community Enterprises division of Community Services. (emphasis added)

An example of social capital that provides tangible and intangible resources is seen in Andrew Muir (Schwab South Africa, Environment/Youth; www.schwabfound.org/content/andrew-muir), who gained reputational resources and inherited an SE from a renowned conversation icon, which paved his way as a social entrepreneur. As stated in his Schwab profile:

[Andrew] dedicated his life to conservation and social development. He was mentored by conservation icon Dr Ian Player for 13 years, and took over his legacy in the management of the various organizations that Player had founded, including the world famous Wilderness Leadership School and Wilderness Foundation. (emphasis added)

Higher Education

Our data show that higher education is an important factor that drives the emergence of SE. Our study demonstrates that higher education is neutral in terms of the (hybrid) logic that it supports. One possible interpretation is that one’s education level—more than one’s specific degree, whether in business or medicine another field—makes individuals abler to recognize social issues and offer solutions. In our study, we find that social entrepreneurs have undergraduate (the majority), master’s, and/or doctoral degrees, with no dominance in specific domain areas. An example of the role of higher education in SE can be seen in Harish Hande’s biography (Schwab India, Energy/Environment/Rural Development; www.schwabfound.org/content/harish-hande):

[Harish] earned his doctorate in energy engineering at the University of Massachusetts, specializing in solar energy. Hande originally started his PhD thesis in heat transfer, but changed his academic focus after visiting the Dominican Republic and observing areas with poverty worse than India using solar energy. Upon returning to Massachusetts, he abandoned his heat transfer thesis and started anew on solar electrification in rural areas, conducting much of his research in India, Sri Lanka and the Dominican Republic. (emphasis added)

Principal Component Analysis

As a follow-up to the qualitative data analysis, we conducted quantitative data analysis to validate the antecedents of SE emergence (as in Fig. 1). We hired two top undergraduate students who had taken a semester-long SE course to independently code Ashoka biographical profiles. We focused on Ashoka only because it provides more detailed biographies than the Schwab thus allowing a higher quality data analysis. Prior to the actual coding, we trained the two coders using a coding scheme derived from Fig. 1 (see Web Appendix 3). The coding scheme comprises multi-item questions that measure the presence or absence of the antecedents or variables (using binary, or 1 and 0 measures, respectively) (Krippendorff 2004). After two training sessions, the two independent coders were able to achieve high levels of accuracy in coding. Following this, the actual coding started and was completed in one week, where 117 biographies of randomly selected Ashoka samples were coded. The inter-coder reliability was 85% with 20% profiles overlap between the two coders.

We conducted principal component analyses (or PCA; Lê et al. 2008; Sebastien et al. 2008) to see whether the eight antecedents of SE emergence (from Fig. 1) could be reduced to a smaller set of dimensions. Prior to the actual analysis, we conducted assumptions testing in SPSS and found all assumptions were met: the data had a normal distribution, the sample size was adequate given the eight antecedents analyzed, no serious outliers in the data, and a sizeable correlation among the eight antecedents existed (Coakes and Steed 2001). The results also showed that Bartlett’s test of sphericity is significant (p < 0.001), and the Kaiser–Meyer–Olkin measure of sampling adequacy met the minimum of threshold—thus it was appropriate to proceed with the PCA. We also performed descriptive statistics analysis by examining the means, standard deviation, and correlation among the eight antecedents (see Web Appendix 4). The PCA analyses showed that eigenvalues dropped significantly after two components were considered, with a cumulative percentage of variance explained for component 1 (27.2%) and 2 (19.1%) at 46.3%; which shows that the eight antecedents can be safely combined into two broader components. Specifically, altruism, spiritualism, collectivism, and ideologism loaded as one component (with rotated component matrix between 0.58 and 0.76) and the entrepreneurialism, professionalism, resources, and higher education loaded as another component (with rotated component matrix between 0.36 and 0.83). In other words, the PCA analysis revealed that altruism, spiritualism, collectivism, and ideologism can be combined into the social skills component; and entrepreneurialism, professionalism, resources, and higher education can be combined into the economic skills component—which essentially confirmed the manual coding results conducted earlier by the authors (see Fig. 1).

To enhance the PCA analysis, we created visualizations of the eight antecedents along the two components using FactoMineR (Lê et al. 2008; Sebastien et al. 2008) and factoextra (Kassambara 2015) packages in R software (see Fig. 2). Figure 2 depicts the arrows with different length,Footnote 4 positionFootnote 5 in the X- and Y-axis, and colorFootnote 6 for each of the eight antecedents. Figure 2 shows that the highest contributors to the social skills component (near the 3 o’clock position), in respective order, are collectivism (29.74%), ideologism (23.78%), altruism (21.12%), and spiritualism (12.94%). The highest contributors to the economic skills component (near 1 o’clock position), in respective order, are entrepreneurialism (44.96%), resources (16.61%), professionalism (14.50%), and higher education (8.72%). Collectivism has the highest and positive correlation with the social skills component (as indicated by its long arrow and position at the right side of the zero point in Y-axis). Likewise, entrepreneurialism has the highest and positive correlation with the economic skills component (as indicated by its long arrow and position above the zero point in X-axis). Overall, these reveal a relatively equivalent share of importance among the top three antecedents in the social skills component and collectivism being a prominent element of the social skills component compared to a disproportionately high share of importance of entrepreneurialism among other antecedents in the economic skills component.

Fig. 2
figure 2

Variable factor map (principal component analysis)

Full size image

To deepen our understanding of how each social entrepreneur integrated the social skills and economic skills components in SE, we created a plot of the principal component scores for the 117 social entrepreneurs on the social and economic skills components (see Fig. 3). As shown in Fig. 3, social entrepreneurs are highly heterogeneous in their background and that there may be no ideal or typical background of a social entrepreneur. There are four groups of social entrepreneurs that can be observed in Fig. 3 and each of them consists of about one-fourth of the sample. One group of SE, located at the northeast position, is called the “social-elite” as they have high social skills and economic skills (e.g., Karen Tse, a US lawyer and long-time social activist whose parents migrated from Hong Kong who started an SE to reform criminal justice system, www.ashoka.org/en/fellow/karen-tse; and Johaness Lindner, an Austrian entrepreneur with a passion for social causes who started entrepreneurship education SE for the deprived, www.ashoka.org/en/fellow/johannes-lindner). Another group of SE, located at the southeast position, is called the “social-grassroots” in that they have strong social skills but lower economic skills (e.g., Adina Bar-Shalom, daughter of Israel’s spiritual leader and an avid social activist who, after all her children were married, started a professional training SE for ultra-Orthodox women, www.ashoka.org/en/fellow/adina-bar-shalom; Charles Banda, a Malawian firefighter and taxi driver and son of a pastor used his retirement time and personal savings to start a SE that provides freshwater to communities, www.ashoka.org/en/fellow/charles-banda).

Fig. 3
figure 3

Individual social entrepreneur plots (principal component analysis)

Full size image

Another group of SE, located at the northwest position, is called the “elite” as they have strong economic skills but lower social skills (e.g., Jack Sim, a successful Singaporean construction entrepreneur who had an emotional turning point in life and started a global toilet movement SE, www.ashoka.org/en/fellow/jack-sim; Mohammad Al-Ubaydli, a British entrepreneur and medical doctor trained at Cambridge who suffered from a rare genetic disease and later started a healthcare SE, http://uk.ashoka.org/fellow/mohammad-al-ubaydli). The last group of SE, located at the southwest position, is called the “commoner” as they have relatively weaker social and economic skills than the other three groups (e.g., Bhargavi Davar, a lower-middle class Indian whose mother suffered mental illness and whose daughter was born with multiple defects and later started a mental health SE, http://india.ashoka.org/fellow/bhargavi-davar; Jessica Mayberry, a middle-class New Yorker TV journalist who due to her disappointment about how news are delivered created a SE to allow marginalized people to become community producers).

Discussion

Prior research merely describes the existence of and tension between social welfare and commercial logics as a central discourse in social enterprises (SE) as hybrid organizations (Cooney 2006; Dart 2004; Doherty et al. 2014). However, little is known about what constitutes each of these logics and why and how individuals are willing and able to combine these contradictory, multiple logics. Our study unpacks and extends the SE-as-hybrid-organization discourse by identifying eight biographical antecedents of the emergence of SE in the form of hybrid organizations (in descending order of importance for each component): collectivism, ideologism, altruism, and spiritualism, which can be aggregated as social skills component; and entrepreneurialism, resources, professionalism, and higher education, which can be aggregated as economic skills component. The integration of the social skills and economic skills highlights the hybrid nature of SEs. Our findings offer finer explanations that go beyond the conventional push or pull theorization in SE (e.g., seeing/experiencing a social problem vs. career development; Yiu et al. 2014; Yitshaki and Kropp 2016) or the motivation for SE (e.g., social responsibility, dissatisfaction with status quo, Christopoulos and Vogl 2015; personal fulfillment, achievement, helping society, Germak and Robinson 2014; altruism or compassion, Miller et al. 2012) and prior experience (Lee and Battilana 2013). We deconstructed hybridity in SE in a more comprehensive manner (i.e., the eight antecedents and their two groupings above) and contributed several new antecedents to the SE literature, which are: (1) collectivism, where SE emerges from collective values and the willingness to collaborate with other groups to improve society; (2) ideologism, where SE evolves as an extension of people’s prior (socio-political) ideology work; (3) spiritualism, in that SE evolves from religious beliefs, contacts with a religious group, or religiously inspired community work; and (4) values and resources that enable the economic skills, which are entrepreneurialism, resources, and professional experience.

Second, our identification of social skills and economic skills as antecedents of SE resonate with the micro-foundations of field-level organizational change literature or the neo-institutionalism theory (Suddaby et al. 2016). Social skills in neo-institutionalism refer to individuals’ sensitivity to the inter-subjective relationships of people in social structures, a cognitive capacity for reading people and the environments, or a general awareness of the defining social order that one is embedded (Fligstein 1997; Suddaby et al. 2016). Our findings demonstrate that people’s “sensitivity” to societal problems (e.g., social injustice, environmental degradation) needs to be better understood from their collectivism, ideologism, altruism, and spiritualism values and resources, as they influence one’s social skills capacity. Thus these four antecedents enrich and deepen the meaning social skills in the neo-institutional theory. As theorized by neo-institutionalists, social position or people’s different social standing in social fields and their awareness of their social standing will have differential effects on their willingness and ability to access key resources (e.g., social capital, status, power, and expertise) to enact social change. The “economic skills” identified in this study (i.e., entrepreneurialism, resources, professionalism, and higher education) can be regarded as a proxy of what neo-institutional theorists call “social position” (Battilana, 2006; Clemens and Cook 1999; Suddaby et al. 2016), which incentivize, enable, and legitimize individuals to enact social change and bridge diverse stakeholders to enable social change. Battilana (2006) argues that social position such as one’s privilege or elite status (that often confers power and resources) is an important factor in understanding why and how people can overcome constraints in their institutional environment. Our study suggests that social position can be better understood by considering individuals’ mix of entrepreneurialism, resources, professionalism, and higher education (rather than privilege/elite status alone).

Our study allows us to develop a typology—based on social skills and social positionFootnote 7—that summarizes the different types of social entrepreneurs based on their hybridity (see Fig. 4). As shown in Fig. 4, our study is the first to demonstrate that social entrepreneurs are not a homogeneous group but are highly heterogeneous in their mix of social skills and social position. This allows us to theorize four types of SE-as-hybrid-organization and their representative examples: (1) the social elite, such as a successful lawyer/entrepreneur who has a long-term passion for social cause or charitable work; (2) the social-grassroots, such as a highly spiritual social activist or firefighter with a long-term passion to improve society; (3) the elite, such as a successful for-profit entrepreneur who faced a life crisis or has illness; and (4) the commoner, such as a lower middle-class mother who had a very sick child and was not satisfied with the status quo and thus created her own solution through SE. These analyses show that combining multiple institutional logics might not be as challenging as presumed in the literature because social skills, a key component in establishing SE, may (1) already exist early in an individual’s life as a propensity to improve others’ welfare regardless of one’s social position, or (2) emerge at some point in one’s life through important life events and experiences at a time when such individual has possessed favorable social position. This finding helps resolve the puzzle about why an individual is willing and or capable of combining multiple institutional logics.

Fig. 4
figure 4

Typology of social enterprise hybridity

Full size image

As Fig. 4 shows, the meaning of SE’s hybrid-ness is richer and more complex than those portrayed in the literature. Moreover, little is known about why and how individuals are able and/or willing to combine multiple (welfare vs. commercial) institutional logics. Our study demonstrates four meanings or types of hybridity in SE, suggesting that there is no “ideal” or typical type of social entrepreneurs as such entrepreneurs can come from individuals with any mix of social skills and social position. We are also the first to show the pluralistic origin of hybrid-ness in SE in that individuals can mix different social skills and social position to start SE and create social change. Interestingly, we did not observe any “highly marginalized/oppressed” individuals became a social entrepreneur. One possible reason is that highly marginalized people might not have sufficient social skills and social position to enact social change, but this becomes possible at a later stage in their life when their social and economic skills have improved to a certain threshold or satisfactory level.

Third, SE scholars tend to dichotomize the “heroic” and “community/collective” representation of SE as agents of change (e.g., Bacq et al. 2016; Montgomery et al. 2012; Nicholls 2010), but our study paints a more complex picture than this. Our scrutiny of the biographical narratives data of Ashoka and Schwab reveals a clear inclusion of both “heroic”- and “community”-driven portrayal of most of the social entrepreneurs where the collective (“community”) and entrepreneurial (“heroic”) nature of social entrepreneurs (see the "Qualitative Findings" section) were clearly present in the biographical narratives; and both collectivism and entrepreneurialism were the biggest contributors to the variance in the data (see Fig. 2). Using our “the elite” SE as an example (Fig. 4), this type of SE may portray a heroic representation of SE. However, “the social-grassroots” and the “the commoner” SEs tend to portray a less heroic and thus more collective/community-driven approach in starting SE. In light of this, we question the usefulness of the heroic versus community dichotomy on SE representation and suggest scholars to devote attention to more promising avenues of research (e.g., the social and economic performance implications of social entrepreneurs from different types of hybridity).

In terms of methodological contributions, this study is one of the first to examine the biographies of a relatively large sample (n = 317) of social entrepreneurs using the established biographical databases of two of the world’s largest SE support organizations: Ashoka and the Schwab Foundation. This addresses the limitations of prior research on SE emergence, which tends to lack larger qualitative studies. Moreover, this study demonstrates the use of mixed methods research that combines Gioia’s methodology with the aid of CAQDAS and principal component analysis—to complement prior studies that predominantly investigated small samples or case studies (e.g., Datta and Gailey 2012; Ruebottom 2013).

Implications

We underscore two important implications. First, the social skills and social position components of the emergence of SE and their eight antecedents that we extracted may provide some clues about the different pathways and factors for people to pursue SE based on certain biographical mix (i.e., collective action experience, ideological work, spiritual engagement, and entrepreneurial experience). Our findings may be useful for policymakers and social impact investors who are in need of profiling tools to classify and select potential social entrepreneurs. Using our findings, such investors can make more informed decisions to invest in certain individuals who are seeking SE funding opportunities. Our finding may also help others understand the gaps within teams of potential social entrepreneurs by suggesting other individuals who might be recruited to strengthen a founding SE team. For would-be entrepreneurs, these antecedents can be used as benchmarks to rate their backgrounds and suitability as social entrepreneurs.

Second, the study may offer new insights into curriculum development for educational institutions that want to foster SE, particularly by pinpointing areas of intervention in the curriculum. For instance, the importance of collectivism (e.g., collective action experience) in the business and non-profit sectors may imply the need to expose students to different business and social organizational skills and models in different sectors in an SE curriculum. It may also imply that different types of activism, religiosity and field experience that enhance students’ social sensitivity can be integrated in SE curriculum. Much more can be explored in terms of designing a curriculum specifically for SE courses; we hope that our study can offer some guidance.

Limitations and Future Research

This study is not without limitations. First, as data sources, the Ashoka and Schwab fellows may be viewed as presenting a bias toward high performers and those who subscribe to the Ashoka and Schwab’s selection criteria; therefore, our findings might only capture the imprint and traits of social entrepreneurs of a certain caliber or with good reputations. Although we cannot fully discount this possible bias in our sampling strategy, the sample size presents a variety of social entrepreneurs with varying performance levels, although none of them can be considered an outright failure. The fellows’ profiles may also be written to support Ashoka’s and Schwab’s goals, and thus there may be a certain political prerogative in the narratives. These biographical profiles are the fellows’ narratives; however, they are simultaneously the narratives of Ashoka and Schwab in that they are written following a format that is prescribed/determined by the support organization. As such, we are arguably analyzing narratives that are co-created by the fellows and the foundations. As with any narrative, these biographical profiles contain certain ways of understanding and explaining the world that reflect the vision and values of their co-creators. Nevertheless, we addressed this possible bias by collecting external sources of information on each of the fellows and can confidently claim that there was a high degree of match between the narratives presented by Ashoka and Schwab and the external information. Future research can compare these fellows with other SE fellows who are members of other SE support organizations (e.g., Skoll, Echoing Green) or SEs not supported by any large SE support organizations and test if they are also driven by the eight antecedents that found in this study. Moreover, our study focused on the biographical antecedents of SE only and although biographical data often contain organizational and institutional antecedents we did not delve deep into organizational and institutional antecedents. Therefore, future research can expand the scope of this study by examining multi-level antecedents at the individual (e.g., biographical), organizational (e.g., type resources and partners) and institutional (e.g., norms, tradition, culture) level. In addition, depending on the institutional settings, an individual’s social position can be embedded in a particular race or ethnicity (e.g., upper class Anglo-Saxon or privileged Chinese Singaporean), which may play a role in the willingness and ability to start SEs. Therefore, why and how race and ethnicity might influence SE emergence is an important question for future research.

Second, some might argue that Ashoka and Schwab are likely to be “individual” than “collective/team based” social entrepreneurs (examples of the latter include Kiva, Marine Stewardship Council and Habitat for Humanity, Montgomery et al. 2012). Our study did not seek to compare and contrast the individual versus collective SE models, and therefore we did not include the latter in our samples. This, however, leaves ample avenues for future research to explore and test the distinctiveness of the antecedents of individual and collective models of SE.

Third, our study does not offer a causal mechanism or explain how the antecedents interact with one another in a way that leads to the emergence of a social enterprise. This shortcoming may suggest an avenue for future research, although we understand that equifinality (that different starting points, e.g., life history, may lead to the same outcome) issues may make such research very difficult. We aimed to take a small preliminary step to map the different antecedents of the emergence of a social enterprise, something that has rarely been studied at the micro-foundational level using well-developed biographical databases with a sample size as large as ours. Future research can present new hypotheses based on our findings and statistically test them using a large-scale survey administered to all Ashoka and Schwab fellows to achieve higher external validity.

Fourth, our study does not examine the “traits” (psychology-centric view) of social entrepreneurs (e.g., risk taking, needs for achievement, locus of control, etc.) or their cognitive logic (e.g., effectuation vs. causation; Sarasvathy 2001). These are promising avenues for future research that complement and extend the narrative approach such as ours that captures the micro (individual)-, meso (organizational)- and macro (institutional)-level factors. Future research can employ experiments and large-scale surveys to understand the influence of the individual’s traits and cognition on behavioral outcomes.

Finally, future research on the antecedents of the emergence of SE can tap into the social skills and social position components and their eight dimensions to examine how they predict the SE performance, learning, social impact, and rhetorical strategies as well as the possible interactions among them.