1 Introduction

A nation’s culture, defined as the shared values, norms, and practices of a society (House et al., 2004), shapes the legitimacy of entrepreneurial activity (Chua et al., 2015; Stephan & Uhlaner, 2010), and what is valued in a specific country impacts the type and number of new ventures that emerge (Liñán et al., 2016; Stephan, 2022; Stephan & Pathak, 2016). In empirical research, culture is conceptualized either as values, which reflect shared ideals of a culture (“how it should be”), or practices, which reflect shared beliefs of how individuals routinely behave in a culture (“how it is;” Hofstede, 2010; House et al., 2004). With a few notable exceptions (e.g., Autio et al., 2013; Hechavarria & Brieger, 2022; Stephan et al., 2015; Thai & Turkina, 2014), entrepreneurship scholars tend to operationalize culture through values (e.g., Bullough et al., 2017; Calza et al., 2020; Canestrino et al., 2020; Jaén et al., 2017; Lee & Kelly, 2019; Morales et al., 2019; Stephan & Pathak, 2016). This perspective is built upon the assumption that the relationship between values and practices is self-reinforcing, such that individuals will act in a manner consistent with their values (Hofstede, 2001; Patterson, 2014).

However, research has revealed a negative relationship between values and practices (Brewer & Venaik, 2010; Maseland & Van Hoorn, 2009; Taras et al., 2010), which suggests that each is likely to relate differently to distinct behaviors due to theoretical and empirical inequivalence (Javidan et al., 2006). Societal inconsistencies between values and practices have critical implications for entrepreneurs, as any dissonance between the two can create uncertainty concerning expected behaviors. An entrepreneur may consider engaging in social entrepreneurial activity in a society that purports to value altruism and generosity. But if in practice, individuals prioritize self-interests and financial success, this misalignment has the potential to make social venture creation less attractive due to distorted perceptions of values-guided behaviors.

The primary purpose of this paper is to operationalize the misalignment between cultural values and practices, which we define as cultural dissonance, and explore the value of this conceptualization by examining how cultural dissonance influences engagement in social entrepreneurial activity. Our theorizing draws on role congruity theory, which explains that prevailing norms and expectations exist concerning how individuals should behave, but individuals often find incongruities between these expectations and their identities (Biddle, 1986; Eagly & Wood, 2012). For instance, some scholars show a negative relationship between uncertainty avoidance values and social entrepreneurship (Canestrino et al., 2020), but others demonstrate a positive relationship between uncertainty avoidance practices and social entrepreneurship (Hechavarría and Brieger, 2022). Moreover, entrepreneurship is a gendered activity (Ahl & Marlow, 2012; Gupta et al., 2009; Jennings & Brush, 2013), and research has established that cultural factors have an important influence on women’s entrepreneurship (Gimenez-Jimenez et al., 2022; Hechavarría and Brieger, 2022). We propose that cultural dissonance will influence prospective male and female social entrepreneurs differently based on congruence or incongruence with overarching cultural factors. Using multilevel logistic regression, we test our hypotheses in a pooled sample of entrepreneurs (N = 23,828).

This research offers two important contributions to the literature. First, we find support for societal misalignment between cultural values and practices, i.e., cultural dissonance, which contributes to recent research calling for an exploration of how cultural factors influence entrepreneurship (Bullough et al., 2022). By showing that cultural dissonance influences social entrepreneurial activity, we inform conversations concerning how cultural attributes shape social entrepreneurship (e.g., Canestrino et al., 2020; Gupta et al., 2022 Hechavarría and Brieger, 2022; Kedmenec & Strašek, 2017; Saebi et al., 2019) and the effects of role (in)congruity (e.g., Anglin et al., 2022a; Cowden et al., 2021; Martiarena, 2022; Yang et al., 2020). Second, we reveal that female entrepreneurs are less sensitive to cultural dissonance than males, thus extending recent discussions on the influence of gender in entrepreneurship (e.g., Alsos & Ljunggren, 2017; Bullough et al., 2017; Hechavarría and Brieger, 2022; Jennings & Brush, 2013; Malmström et al., 2017; Malmström et al., 2020; Yang & del Carmen Triana, 2019). We propose entrepreneurial activity is driven, at least in part, by whether an entrepreneur views his or her self-concept as congruent or incongruent with higher-order dimensions of culture and gender roles.

2 Theoretical background

2.1 Culture and social entrepreneurship

Culture operates as a generative scheme invoked by entrepreneurs as a tool to inform their actions (Bruni et al., 2004; Davidsson, 1995; Lauring et al., 2018; Patterson, 2014; Swidler, 2008) and sets boundaries for entrepreneurial activity (Liu & Almor, 2016; Muralidharan & Pathak, 2017). A favorable national culture can create an advantageous institutional environment which makes entrepreneurship socially valued and recognized (Davidsson, 1995; Mueller & Thomas, 2001). While economic entrepreneurship is focused on the exploitation of opportunities for personal or stakeholder wealth through financial pursuits, social entrepreneurship centers around engaging in altruistic, ecological, or communal opportunities through social provisioning, and focuses on activities motivated by societal needs (Austin et al., 2006; Zahra et al., 2009). Both cultural values and cultural practices have been established as important to entrepreneurial intentions and action (e.g., Canestrino et al., 2020; Jaén et al., 2017; Kedmenec and Strašek, 2017; Puumalainen et al., 2015; Short et al., 2009; Stephan et al., 2015).

Cultural values refer to the enduring beliefs, attitudes, and principles that guide behavior within a particular culture (Hofstede, 1980). These values can influence a society’s attitude towards social entrepreneurs, as well as the likelihood of individuals within that society engaging in social entrepreneurship. For example, collectivist cultures, which value interdependence and community harmony, are more likely to support social entrepreneurship, but individualistic cultures, which prioritize individual achievement and autonomy, may be less supportive of social entrepreneurship. Cultural values can not only shape how business is conducted within new ventures but also reveal how and why venture creation is motivated as a predecessor to action (Hayton et al., 2002). Not all societies foster entrepreneurial action with equal effectiveness (Bruton et al., 2010; De Clercq et al., 2013), but through cultural values, can either encourage or discourage entrepreneurial activity (Hayton et al., 2002).

Cultural practices refer to the shared behaviors and customs that are specific to a particular culture (Patterson, 2014; Therborn, 2002). Cultures with a strong tradition of philanthropy, voluntary service, a history of social activism, and community organizing are more conducive to the development of social entrepreneurship (Bacq et al., 2013). Scholars argue that cultural practices guide behaviors through a “dominant logic of action” which reflects a repertoire of habits, skills, and styles (Swindler, 1986). Thus, cultural practices are argued to be a better measurement of culture than values, because individuals are likely to conform to societal norms by repeating typical behaviors (Powell et al., 2012; Stephan et al., 2015). This approach has been used to explain women’s entrepreneurship (Bullough et al., 2017; Hechavarría and Brieger, 2022), as well as cross-cultural differences in economic and social entrepreneurship (Autio et al., 2013; Stephan & Uhlaner, 2010; Thai & Turkina, 2014).

To date, research exploring the link between specific cultural dimensions and social entrepreneurial activity offers mixed results. A recent paper builds on the literature exploring female entrepreneurship (Urbano Pulido et al., 2014) to reveal that female entrepreneurs are more likely to engage in social entrepreneurship when the practices of power distance, humane orientation, and in-group collectivism are low, and the practices of future orientation and uncertainty avoidance are high (Hechavarría and Brieger, 2022). Differently, power distance, individualism, and uncertainty avoidance are all negatively related to traditional (commercial) entrepreneurship and power distance is negatively related to social entrepreneurship (Puumalainen et al., 2015). Other scholars use values to explore the contextual drivers of different types of entrepreneurial activity (e.g., Kedmenec & Strašek, 2017), but after finding only partial support for certain dimensions, scholars concluded that culture was not a sufficient justification for national differences in social entrepreneurship rates (Canestrino et al., 2020). Interestingly, however, other studies have revealed that specific dimensions interact with demographic categories (i.e., age, education) to predict social entrepreneurial activity (Lee & Kelly, 2019; Urbano Pulido et al., 2014). A summary of research testing the relationship between cultural factors and social entrepreneurship is presented in Table 1.

Table 1 Cultural factors and social entrepreneurship
Full size table

2.2 Cultural dissonance

Culture is a construct that is often represented using an “onion” metaphor, wherein cultural values are at the core, with practices making up the outer layer (Hofstede, 2001). The GLOBE study contains separate measures and offers a unique opportunity for scholars to explore differences between these cultural facets. Cultural consonance reflects the degree of coherence an individual experiences within a specific culture (Dressler et al., 2018; François Dengah et al., 2022). A central assumption of cultural consonance revolves around how individuals understand patterns of agreement between norms and behaviors within that culture (Romney et al., 1986).

But evidence suggests that cultural values and practices are negatively correlated, and each relates differently to distinct behaviors (House et al., 2004; Javidan et al., 2006). Though some scholars have argued that differences between values and practices might lie in measurement concerns (e.g., Maseland & Van Hoorn, 2009), others suggest that substantial cultural shifts or different motivational factors could drive inconsistencies (Cohen, 2001; Taras et al., 2010). For example, in organizations, values such as, “employees should not steal from the company,” may deviate substantially from practices, i.e., “employees do steal from the company,” with disparate behaviors (i.e., stealing vs. not stealing; Lea, 1999). Accordingly, we introduce the concept of cultural dissonance. Low cultural dissonance reflects consistency between values and practices; behaviors are equally desirable and practiced. High cultural dissonance reflects inconsistency between values and practices; behaviors are not equally desirable and practiced. We expect that cultural dissonance will play a role in explaining the choice to engage in social entrepreneurial activity. Further, it is recommended to consider each dimension separately in order to fully understand plausible explanations for dissonance (Brewer & Venaik, 2010), which we do in greater detail, below.

A socially supportive culture (SSC) reflects a positive societal climate in which individuals support each other, are generous and caring, have high levels of social capital, and are characterized by a high positive loading of humane orientation and a low positive loading of assertiveness (Stephan & Uhlaner, 2010). Humane orientation is the “degree to which a collective encourages and rewards individuals for being fair, altruistic, friendly, generous, caring, and kind to others” (House et al., 2004: 30). Assertiveness is the “degree to which individuals are assertive, confrontational, and aggressive” (House et al., 2004: 30). SSC cultural dissonance manifests when inconsistences viewed at the societal level create ambiguity around behavioral legitimacy, related directly to the importance of cooperation, friendliness, supportiveness, cooperation, social capital, and social relationships.

A performance-based culture (PBC) reflects a societal climate in which individual accomplishments and financial performance are rewarded and systematic, future-oriented planning is viewed as necessary to achieve success (Stephan & Uhlaner, 2010). PBCs are characterized by a high positive loading of future orientation, uncertainty avoidance, and performance orientation, and a high negative loading of in-group collectivism and power distance. Future orientation is the “extent to which individuals engage in future-oriented behaviors, such as delaying gratification, planning, and investing in the future” (House et al., 2004: 30). Uncertainty avoidance is the “extent to which a society, organization, or group relies on social norms, rules, and procedures to alleviate unpredictability of future events” (House et al., 2004: 30). Performance orientation is the “degree to which a collective encourages and rewards group members for performance improvement and excellence” (House et al., 2004: 30). In-group collectivism is “the degree to which individuals express pride, loyalty, and cohesiveness in their organizations or families” (House et al., 2004: 30). Power distance is the “degree to which members of a collective expect power to be distributed equally” (House et al., 2004: 30). PBC cultural dissonance manifests when inconsistences viewed at the societal level create ambiguity around planning ahead, an objective view of rules and regulations, performance-based appraisal systems, the importance of personal relationships, and unequal distributions of power.

2.3 Role congruity theory

Role congruity theory is grounded in social role theory, which submits that prevailing norms, expectations, and stereotypes (i.e., social roles) exist concerning how groups of individuals should behave (Biddle, 1979, 1986; Eagly & Wood, 2012). Social roles are socially shared behavioral expectations that attach to an individual’s position as it relates to a particular social setting or social category (Biddle, 1979). Conforming to behavioral expectations drives evaluations by others, in that acting in accordance with prescribed roles results in favorable evaluations but acting incongruent with these roles can elicit social sanctions (Bolino & Turnley, 2003; Eagly & Karau, 2002).

One of the key predictions of role congruity theory is that individuals who experience role incongruity are less likely to engage in activities associated with their incongruent role (Eagly & Karau, 2002). This implies that individuals whose personal characteristics or roles are incongruent with societal expectations of entrepreneurs may be less likely to engage in entrepreneurial activity. The role of the economic entrepreneur tends to be associated with the characteristics of masculinity, aggressive risk-taking, and overconfidence (Anglin et al., 2018; Balachandra et al., 2019; Estrin et al., 2013; Jonason & Fletcher, 2018), factors which are often attributed to male-based stereotypes. Entrepreneurs that perceive congruity with these characteristics are likely to engage in entrepreneurial action in cultures that reflect these factors. However, social entrepreneurs are more often associated with compassion, empathy, agreeableness, and concern for others (Nga & Shamuganathan, 2010; Pan et al., 2019; Saebi et al., 2019), factors which are often attributed to female-based stereotypes. Social entrepreneurs that perceive congruity with these characteristics are likely to engage in social entrepreneurial action in cultures that reflect these factors (Stephan et al., 2015; Westlund & Adam, 2010). Divergence between cultural values and cultural practices could act as a facilitator to action as social entrepreneurs function as agents of change, allowing or encouraging realignment of cultural values and practices via entrepreneurial activities. For instance, launching a new venture with a social mission in a culture with a high level of SSC cultural dissonance might encourage practices reflecting humane orientation, thus decreasing the level of dissonance between values and practices. However, cultural dissonance also has the potential to impede social action if interpreted by social entrepreneurs in such a way that they conclude they are illegitimate actors and choose instead to engage in other activities.

3 Hypothesis development

3.1 Socially supportive cultures

Scholars consistently show positive effects of socially supportive cultures on entrepreneurship, explained by decreased risk factors in the venture creation process (Gimenez-Jimenez et al., 2022) and by allowing access to critical resources which are essential for venture survival (Stephan & Uhlaner, 2010). Socially supportive cultures are specifically noted to stimulate social entrepreneurship, as the culture reflects those who value trust, cooperation, and caring behavior (Stephan et al., 2015; Westlund & Adam, 2010). SSC cultural dissonance occurs when inconsistences viewed at the societal level create ambiguity around expected behaviors related to humane orientation and assertiveness. For instance, prospective social entrepreneurs might interpret societal values to indicate that individuals will encourage and reward altruism, social action, and motives addressing social problems, reflecting the expected high positive loading of humane orientation, but discourage assertive pursuit of self-interests and financial rewards, reflecting the expected high negative loading of assertiveness. However, if practiced behaviors do not reflect these values, there is an increased risk of failure for new social ventures because these circumstances are likely to reflect a lower demand for social entrepreneurial activity. We therefore predict that as SSC cultural dissonance increases, engagement in social entrepreneurial activity will decrease.

  • Hypothesis (H1). There is a negative relationship between SSC cultural dissonance and engagement in social entrepreneurial activity.

3.2 Performance-based cultures

Scholars consistently show positive effects of performance-based cultures on economic entrepreneurship, explained through high achievement orientation and an inclination for systematic planning (Brinckmann et al., 2010; Delmar & Shane, 2003; Rauch et al., 2000). Performance-based cultures are specifically noted to stimulate economic entrepreneurship, as this culture reflects those who value financial achievement and systematic business planning, and rewards individuals that engage in similar cultural practices, such as capitalizing on innovation or taking initiative (Laskovaia et al., 2017). Economic entrepreneurship is perceived as socially desirable in performance-based cultures due to norms that encourage and reward individual efforts and accomplishments (Thai & Turkina, 2014). PBC cultural dissonance occurs when inconsistencies viewed at the societal level create ambiguity around expected behaviors related to future orientation, uncertainty avoidance, performance orientation, in-group collectivism, and power distance. That is, prospective entrepreneurs might interpret societal values to indicate that individuals will encourage and reward economics pursuits, systematic future planning, and financial rewards, reflecting the expected high positive loading of future orientation, uncertainty avoidance, and performance orientation, but discourage strong social relationships, power equality, and seeking to solve societal problems, reflecting the expected high negative loading of in-group collectivism and power distance. However, if practiced behaviors do not reflect these values, we predict that prospective social entrepreneurs will be more likely to engage in social entrepreneurial activity because these circumstances are likely to reflect a higher demand for social entrepreneurship through behaviors that are inconsistent with cultural factors, thus encouraging social action (Whatley et al., 1999). We therefore predict that as PBC cultural dissonance increases, engagement in social entrepreneurial activity will also increase.

  • Hypothesis (H2). There is a positive relationship between PBC cultural dissonance and engagement in social entrepreneurial activity.

3.3 Gender role congruity theory

Gender role congruity theory addresses the specific importance of gender roles in behavioral differences (Eagly et al., 2000) in explaining that society sets expectations about which behaviors or characteristics are considered masculine or feminine (Eagly et al., 2000; Fiske, 2000). For example, males are often expected to be masculine and display behaviors that reflect assertiveness and strength, but women are often expected to be feminine and display behaviors that reflect compassion and empathy (Vogel et al., 2003). Gender role congruity theory builds on these stereotypes by comparing beliefs about how men and women should behave with understandings of how men and women do behave, resulting in favorable perceptions when these norms are in alignment (Eagly & Karau, 2002). These stereotypes are often so deeply embedded that they are endorsed by both men and women, which leads to gender bias operating as an unconscious cognitive heuristic in decision making (Heilman, 2001; Rudman & Phelan, 2008).

Gender norms can be viewed as essential aspects of culture (Feldmann et al., 2022). Entrepreneurship is a gendered activity (Ahl & Marlow, 2012; Gupta et al., 2009), generally stereotyped as a male-dominated profession (Jennings & Brush, 2013). However, social entrepreneurs tend to be associated with empathy, compassion, and interest in the well-being of others (Pan et al., 2019; Saebi et al., 2019), characteristics which consistently conform to prevailing female stereotypes. Female entrepreneurs are more likely to pursue social missions, in alignment with perceptions of feminine attributes and social entrepreneurship (Calic & Mosakowski, 2016; Meyskens et al., 2011). We would therefore expect men to be viewed more favorably when pursuing economic entrepreneurship and women to be viewed more favorably when pursuing social entrepreneurship, as each role shares behavioral norms with gender roles typically assigned to men and women (Lee & Huang, 2018).

Consistency with gender roles tends to elicit a positive response from observers, while inconsistency can bring social sanctioning (e.g., Eagly & Karau, 2002; Gupta et al., 2018). The female gender role is stereotypically associated with demonstrating care for others, empathy, and offering social support (Eagly & Karau, 2002; Rosette & Tost, 2010). For example, in times of stress, women are more likely to draw on socially supportive networks to provide support to other (Taylor et al., 2000). Gender-based differences in motivations and outcomes exist within both socially supportive and performance-based cultures. François Dengah and colleagues (2022) found that men were more motivated by their networks than women. SSC cultural dissonance creates ambiguity around expected behaviors related to (high) humane orientation and (low) assertiveness, factors that align with traditional female roles. Because female prospective social entrepreneurs are more likely to self-select into social entrepreneurial activity to maintain gender role congruency between social entrepreneurship and their own internalized understanding of gender, we predict that they will be less affected by SSC cultural dissonance. However, we expect that male prospective social entrepreneurs will perceive gender role incongruency with the role of the social entrepreneur and SSC cultural factors, making them less likely to engage in social entrepreneurial activity.

  • Hypothesis (H3). The negative effects of SSC cultural dissonance on social entrepreneurial activity will be more pronounced for male entrepreneurs than female entrepreneurs.

The male gender role is stereotypically associated with agentic, high-status, and self-oriented behaviors (Anglin et al., 2018; Jonason & Fletcher, 2018). For example, men are viewed as driven by economic and achievement motivations (Ljunggren & Kolvereid, 1996). The role of the economic entrepreneur aligns well with this role, as men tend to be viewed more favorably when new venture planning highlights economic performance, due to the perceived congruity between the gender role and the role of the economic entrepreneur (Anglin et al., 2022a; Malmström et al., 2020). PBC cultural dissonance creates ambiguity around expected behaviors related to (high) future orientation, uncertainty avoidance, and performance orientation, and (low) in-group collectivism and power distance, factors that align with traditional male gender roles. Because male prospective social entrepreneurs will likely be motivated to self-select into traditional entrepreneurial activity to maintain gender role congruency between economic entrepreneurship and their own internalized understanding of gender, we predict that they will be more affected by PBC cultural dissonance. However, we expect that female prospective social entrepreneurs will experience a lesser reaction to PBC cultural dissonance due to perceived gender role congruity with the role of the social entrepreneur.

  • Hypothesis (H4). The positive effects of PBC cultural dissonance on social entrepreneurial activity will be more pronounced for male entrepreneurs than female entrepreneurs.

4 Methodology

4.1 Sample

To generate the dataset, we combined individual-level data on social entrepreneurial activity (SEA) from entrepreneurs in the 2015 Global Entrepreneurship Monitor (GEM) survey with country-level data from the Global Leadership and Organizational Behavior Effectiveness (GLOBE; House et al., 2004) and World Bank (WB). The GEM research program constitutes the single largest and most recognized program to systematically research the prevalence, determinants, and consequences of entrepreneurial activity on an international level (Bergmann & Stephan, 2013). Consistent with research utilizing the GEM, the sample is weighted according to the adult labor force (ages 18–64) for the respective countries provided by GEM. Detailed descriptions of the methods and sampling frame used to generate the GEM database are reported in Reynolds et al. (2005) and the GEM manual (Bosma et al., 2012). All data were combined into one dataset containing the list of countries, their classification as geographical areas, the stage of economic development, and data for each cultural dimension. We selected 1-year time-lagged country-level variables from the WB dataset to reduce the potential for endogeneity between cultural dissonance and the dependent variable, social entrepreneurial activity. Once the data were cleaned and screened for incomplete responses, we identified and removed outliers and careless responses using a LongString indexFootnote 1 (Meade & Craig, 2012). The final sample contained 23,828 entrepreneurs from 33 countries.

4.2 Measures

Consistent with research that utilizes GEM to investigate social entrepreneurial activity (e.g., Bosma et al., 2012; Gras et al., 2014; Hechavarría, 2016; Hechavarría and Ingram, 2019; Hechavarria & Brieger, 2022; Muralidharan & Pathak, 2017), we aggregated the social entrepreneurship questions to represent our dependent variable, engagement in social entrepreneurial activity. The GEM study took a broad view of social entrepreneurship, including ventures with purely social and environmental goals, as well as hybrids, which is consistent with the accepted definition of social entrepreneurship (Dees, 1998, 2018; Mair & Marti, 2006). Respondents were asked a corresponding set of questions about starting and owner-managing “any kind of activity, organization or initiative that has a particularly social, environmental or community objective” to be identified as social entrepreneurs (see, Mair & Marti, 2006; Zahra et al., 2009). Following well-established protocols (e.g., Estrin et al., 2013; Stephan et al., 2015), we created one primary indicator to measure the dependent variable, coded = 1 if the individual met the criteria for a nascent or operating social entrepreneur, or = 0 if not.

The GLOBE study assesses cultural dimensions encompassing both values and practices. There are measures for values and practices in each of the following dimensions: power distance, uncertainty avoidance, in-group collectivism, assertiveness, performance orientation, future orientation, and humane orientation (House et al., 2004). We used Stephan and Uhlaner’s (2010) approach to operationalize culture in two higher-order dimensions. SSC consists of an average of two dimensions — humane orientation and assertiveness (reverse scored). PBC consists of an average of five dimensions — future orientation, uncertainty avoidance, performance orientation, in-group collectivism (reverse scored), and power distance (reverse scored).

We measured cultural dissonance for each higher-order dimension on the response bias corrected scores of the GLOBE values and practices (House et al., 2004). For each country, cultural dissonance was calculated as the difference between the value and the practice score on the respective GLOBE dimension. In creating our measure, we used the absolute rather than the squared value of the difference to avoid distortion of regression results, as squaring scores can inappropriately emphasize outliers (Cohen et al., 2013). Gap scores are commonly used in business research (Homburg et al., 2009). Consistent with this approach, we conducted an additional robustness check. To address the possibility of low reliabilities, which may occur when component scores have low reliabilities and are positively correlated (Edwards, 2001), we tested the correlations and differences in means between the GLOBE values and practices in our sample. We found statistically significant negative correlations between values and practices on all seven cultural dimensions and between the two second-order dimensions as calculated for socially supportive and performance-based cultures; therefore, there is little concern for low reliability in our data. Using t-tests, we found evidence for significant differences between values and practices at the level of each cultural dimension. These results are presented in Table 2, and means comparisons are presented in Fig. 1. Finally, gender was captured from the GEM survey data, where women = 1 and men = 0.

Table 2 Value-practice misalignment
Full size table
Fig. 1
figure 1

Values-practice misalignment. HO, humane orientation; UA, uncertainty avoidance; PD, power distance; AS, assertiveness; PO, performance orientation; FO, future orientation; IC, in-group collectivism

Full size image

At the individual level, we included several control variables often identified as factors influencing entrepreneurial activity: age, education, and employment status, and personal characteristics found to influence the propensity of an individual to engage in entrepreneurship: start-up skills, fear of failure, knowing a current entrepreneur, and opportunity identification (Brieger & Clercq, 2019; Estrin et al., 2013; Nga & Shamuganathan, 2010). At the national level and consistent with cultural research in entrepreneurship (Stephan, 2022), we controlled for national wealth, measured as gross domestic product (GDP per capita) and GDP growth, as both relate negatively to national entrepreneurship rates (e.g., Minniti et al., 2006; Wennekers et al., 2005). Finally, we controlled for unemployment rate, country population size, and population growth, as each has been linked to entrepreneurial activity (Wennekers et al., 2005). All variables and data sources are presented in Table 3.

Table 3 Variables and data sources
Full size table

4.3 Analysis

We tested our hypotheses by fitting a series of logistic multilevel regression models in which individuals (level 1) were nested within countries (level 2). Because the data are nested within countries (Guo & Zhao, 2000), multilevel logistic regression is the most appropriate statistical method. Multilevel modeling is preferred over conventional single-level regression analysis for several reasons. First, multilevel modeling reduces the potential for type I errors that might occur through not acknowledging the existence of higher-level (national) data. Instead, multilevel modeling provides a systematic and unbiased analysis of how variables and/or the interactions between variables measured at various levels of a hierarchical structure affect the dependent variable. Second, multilevel modeling corrects for bias that may result from clustering, i.e., non-independence of observations within the same country. Finally, multilevel modeling provides robust standard errors, confidence intervals, and significance tests. Our approach is consistent with recommendations and recent usage of multilevel analysis in entrepreneurship research (e.g., Autio et al., 2013; Estrin et al., 2013; Lepoutre et al., 2013; Stephan et al., 2015; Yang et al., 2020).

To estimate the influence of country-level cultural factors (level 2) on individual social entrepreneurial activity (coded as a binary variable), we employed mixed-effects logistic regression that assumes unobserved country-specific effects to be randomly distributed with a mean of zero, constant variance, and uncorrelated to the predictor variables. This allows the constant term to vary randomly across countries (level 2). Using STATA 16, we utilized the melogit command to fit our data with a mixed-effects logistic model. Following prior convention, we proceeded with a three-step estimation strategy to examine the predictors of social entrepreneurship. First, we estimated the between-group (country) variance in the dependent variables by including no predictors in our mixed-effects logistic regression model. This regression model is described as the “null model.” We observed significant country-level variance (0.533), which provides strong statistical support for the choice of a multilevel model over a simple logistic regression model (Snijders & Bosker, 2011). Second, we added gender, then the cultural dissonance predictor variables to test the direct effects on social entrepreneurship engagement. Finally, we added the interaction terms. The reduction in the variance of the random intercept from the null model provides a measure of the extent to which our country-level predictors exclusively account for the proportions of the remaining variance.

To confirm the nature and significance of the interaction effects, we calculated the marginal probabilities using the margins command in STATA 16 to generate average marginal effects using the coefficients generated from the mixed effects melogit regression. To verify the appropriateness of a multilevel regression model, we examined intra-class correlations (ICC; Hox et al., 2017). When ICC shows significant national differences in an individual-level dependent variable, usage of multilevel modeling is indicated. The ICC value in our dataset was 0.139, indicating that 13.9% of the total variance resided at the country level for each social entrepreneurial activity. This is within the normal range for management research and indicates that country-level variance is significant, justifying the application of multilevel regression (Bliese, 2000; Hox et al., 2017). We tested for multicollinearity using the variance correlation estimator (VCE) and found that the highest correlation between any two variables was 0.002. This indicates that multicollinearity is unlikely to be a concern in our data.

5 Results

5.1 Main effects

Table 4 reports descriptive statistics for all variables. Table 5 displays the effects of cultural dissonance on social entrepreneurial activity and the moderating effects of gender. Table 5, model 3, shows a negative relationship between SSC cultural dissonance and social entrepreneurial activity (β =  − 0.514, p = 0.01), supporting hypothesis 1. Table 5, model 5, shows a positive relationship between PBC cultural dissonance and social entrepreneurial activity (β = 0.609, p = 0.01), supporting hypothesis 2.

Table 4 Descriptive statistics
Full size table
Table 5 Multilevel logistic regression effects on social entrepreneurial activity
Full size table

5.2 Moderating effects

Table 5, model 4, shows that the relationship between SSC cultural dissonance and social entrepreneurial activity was moderated by gender (β = 0.349, p = 0.01), supporting hypothesis 3. The baseline effect of being a female entrepreneur in this model was negative (β =  − 0.661, p = 0.001). Therefore, the relative risk ratio comparing male to female entrepreneurs (holding the other variables in the model constant) would be expected to decrease by a factor of 0.627 (odds ratio) when cultural dissonance is zero.

Table 5, model 6, shows that the relationship between PBC cultural dissonance and social entrepreneurial activity was moderated by gender (β =  − 0.481, p = 0.01), supporting hypothesis 4. The baseline effect of being a female entrepreneur in this model was positive (β = 0.428, p = 0.003). Therefore, the relative risk ratio comparing male to female entrepreneurs (holding the other variables in the model constant) would be expected to increase by a factor of 1.656 (odds ratio) when cultural dissonance is zero. Overall, the interaction terms indicate that the effects of gender on social entrepreneurial activity are conditioned on cultural dissonance.

5.3 Robustness checks

Using the coefficients generated in our analysis, we conducted a marginal effects analysis (Busenbark et al., 2022; Chen et al., 2022), which yields the marginal probabilities for SSC and PBC cultural dissonance and social entrepreneurial activity by gender, thus offering additional evidence in support of H3 and H4. The results are presented in Table 6.

Table 6 Marginal effect analysis
Full size table

As interpretation of interaction coefficients from logistic regression is not straightforward (Ai & Norton, 2003) and to further illustrate our findings, we provide interaction plots based on the marginal effects that show the average predicted probability of social entrepreneurial activity at different levels of cultural dissonance for male and female entrepreneurs (Figs. 2 and 3). The vertical axis denotes the probability that an individual engages in social entrepreneurial activity (i.e., p[DV = 1]) while the horizontal axis represents cultural dissonance. The interaction plots provide further support for H3 and H4. As SSC cultural dissonance increases, social entrepreneurial activity decreases, and the effect is less pronounced for female entrepreneurs than male entrepreneurs (H3; Fig. 2). As PBC cultural dissonance increases, social entrepreneurial activity increases, and the effect is more pronounced for male entrepreneurs than female entrepreneurs (H4, Fig. 3). Collectively, as cultural dissonance increases, we find that gender becomes a stronger predictor of social entrepreneurial activity.

Fig. 2
figure 2

Socially supportive culture: cultural dissonance interaction effects

Full size image
Fig. 3
figure 3

Performance-based culture: cultural dissonance interaction effects

Full size image

Finally, consistent with cultural research (e.g., Hechavarría & Ingram, 2019), we conducted one additional robustness check, in which we categorized countries according to competitiveness across stages of economic development: (1) factor-driven (extractive in nature), (2) efficiency-driven (exhibiting scale-intensity), and (3) innovation-driven (characterized by the production of new and unique goods and services using pioneering methods; Acs et al., 2008), to identify whether it was possible to detect subtle variations in effects.

We found additional support for our results, such that the relationship between SSC cultural dissonance and social entrepreneurial activity was moderated by gender in factor-driven economies (β = 0.213, p = 0.03), efficiency-driven economies (β = 0.319, p < 0.01), and innovation-driven economies (β = 0.294, p < 0.02). In performance-based cultures, we found that the relationship between PBC cultural dissonance and social entrepreneurial activity was moderated by gender in factor-driven economies (β =  − 0.398, p = 0.001), efficiency-driven economies (β =  − 0.215, p = 0.001), and innovation-driven economies (β =  − 0.355, p = 0.004).

6 Discussion

The purpose of this study was to explore how misalignment between cultural values and cultural practices, which we define as cultural dissonance, influences engagement in social entrepreneurial activity. Both values and practices are both presumed to be guides of behavior, such that alignment between the two provides stronger guidance for how to behave. However, cultural dissonance is likely to create significant uncertainty concerning desired and expected behaviors in societies. Our findings concerning cultural dissonance have implications for research on cultural attributes and how cultural dissonance and individual characteristics of the entrepreneur influence social entrepreneurial activity.

6.1 Cultural dissonance

One of the most interesting findings from our study is that cultural values and cultural practices are negatively correlated, which has implications for research in terms of how we conceive of and measure culture in the context of entrepreneurship. Research tends to focus on the influence of either cultural values or practices (e.g., Bullough et al., 2017; Canestrino et al., 2020; Kedmenec & Strašek, 2017; Puumalainen et al., 2015; Short et al., 2009). The dominant perspective in the literature is that values and practices are self-reinforcing: individuals behave in a particular way because they hold particular views on “how it should be” (values; Patterson, 2014). Our findings suggest an empirical approach to values and practices as interchangeable conceptualizations of culture will likely result in error (Friedrich, 2010) by failing to account for any potential divergence between the two.

We propose that cultural dissonance is itself a distinct cultural belief, in that acknowledging inconsistencies between societal expectations and behaviors has the potential to offer insight beyond isolated examination of values or practices alone. A significant body of research has focused on investigating which cultural factors facilitate or inhibit social entrepreneurial activity (Canestrino et al., 2020; Jaén et al., 2017; Short et al., 2009; Stephan et al., 2015; Thai and Turkina, 2014). Developing a deeper understanding of how and why cultural values are espoused but not lived creates an opportunity for a more precise investigation of how specific cultural factors shape social entrepreneurial activity, but also into broader societal-level triggers, barriers, and/or explanations for innovation and entrepreneurship.

6.2 Cultural dissonance and social entrepreneurial activity

A further implication of our findings relates to how role (in)congruity with cultural factors viewed at the societal level impacts social entrepreneurial activity. Our research shows that cultural dissonance is a critical consideration for social entrepreneurs. Prior work demonstrates that social entrepreneurship is encouraged in cultures that emphasize activities around social welfare, sustainability, or solving social problems and discouraged in cultures that prioritize individualist rent-seeking and wealth accumulation (e.g., Bacq & Eddleston, 2018; Brieger and Clercq, 2019; Hechavarría, 2016; Stephan et al., 2015), and this assumption holds when cultural values and cultural practices are aligned. However, our analysis reveals that as cultural dissonance increases, social entrepreneurial activity is increasingly impacted.

As a unique conceptualization of culture, we found that the impact of cultural dissonance on social entrepreneurship engagement was based on the underlying cultural dimensions. We examined cultural dissonance within two higher-order dimensions of culture, and the results indicated that the overall effects of cultural dissonance are not necessarily entirely positive or negative, but instead, depend on the nature of the cultural dimension and the overarching context. We contend that values and practices alone are insufficient in explaining cultural effects in certain contexts. Rather, it is recommended that scholars conceptualize and operationalize culture through values, practices, and/or dissonance, based on theoretical grounding of the research question(s) of interest. In the specific context of social entrepreneurship, our research shows that cultural dissonance has a distinct and separate phenomenology that offers deeper insight into social entrepreneurial activity than either values or practices alone.

Further, our conceptualization of cultural dissonance is informed by theory from action-based models of cognitive dissonance, which submit that the greater the magnitude of the dissonance, the more likely it is that an individual will take action to reduce the discrepancy causing the dissonance (Festinger, 1957). Any dissonance has the potential to influence action to some degree (Smith & Semin, 2004), but more immediate, clear, and insistent action implications result from certain dissonant cognitions (Harmon-Jones et al., 2015). Role congruity theory similarly states that individuals experience cognitive and emotional discomfort when their personal characteristics or roles are incongruent with societal expectations (Eagly & Karau, 2002), and has increasingly been applied in the management literature to explain roles, tasks, and choices of individuals based on shared expectations of roles defined by social categories (e.g., del Carmen Triana et al., 2024; Klein et al., 2021) and the psychological discomfort that manifests when individuals act in discordance with these expectations. From this perspective, cultural dissonance could be creating a cognitive response in prospective entrepreneurs due to how they perceive the role of an entrepreneur as (in)congruent with cultural values, practices, and/or the (mis)alignment between the two. Consequently, this dissonance motivates action based on the discomfort experienced due to uncertainty and ambiguity created by cultural dissonance, because behavioral guides from values and practices become less clear and more uncertain as dissonance increases.

This viewpoint offers an explanation for why an individual might choose to engage in social versus economic entrepreneurial activity, thereby extending our understanding of the effects of role (in)congruity on entrepreneurial action (e.g., Anglin et al., 2022a, 2022b; Bullough et al., 2022; Calic et al., 2023; Cowden et al., 2021; Martiarena, 2022; Yang et al., 2020). Consideration of any responses to cultural dissonance offers rich opportunity for future scholars interested in how context influences action, particularly considering recent conversations surrounding contextualization of entrepreneurship (e.g., McMullen et al., 2021; Welter & Baker, 2021; Welter et al., 2019).

6.3 Moderating effects

Our findings have implications for recent conversations in entrepreneurship on the influence of gender (e.g., Alsos & Ljunggren, 2017; Bullough et al., 2017; Butticè et al., 2023; Gimenez-Jimenez et al., 2022; Gupta et al., 2022; Hechavarria & Brieger, 2022; Jennings & Brush, 2013; Malmström et al., 2017; Malmström et al., 2020; Yang & del Carmen Triana, 2019). We demonstrated that male entrepreneurs were more sensitive to cultural dissonance than female entrepreneurs, which suggests that entrepreneurial action is driven, at least in part, by whether an entrepreneur views his or her gender-based self-concept as (in)congruent with specific higher-order dimensions of culture.

An individual’s self-concept consists of three principal elements: “the imagination of our appearance to the other person; the imagination of his judgment of that appearance (i.e., cognitively knowing), and some sort of self-feeling, such as pride or mortification” (Cooley, 1922: 184). The individual’s notion of “the self” is entirely dependent on social responses to contextual factors, such as culture. By operationalizing cultural dissonance, our research highlights how gender manifests in socially and culturally defined prescriptions and beliefs about the behavior of men and women (Anselmi & Law, 1998). As a result, we suggest that the increased sensitivity to cultural dissonance by male social entrepreneurs is a result of psychological discomfort as created by the misalignment with the self-concept, making it more likely that these entrepreneurs will take action to reduce this discomfort. This finding informs research on stereotype threat, which suggests that individuals are hindered in situations where stereotypes suggest that the group to which they belong has weaknesses in a particular domain and helped when the association suggests strength (Gupta et al., 2014). Similarly, expectancy violation theory contends that forms of counter-stereotypical behavior can have positive effects when the unexpected characteristic carries positive connotations (e.g., cooperation) and not negative ones (e.g., dominance; Jussim et al., 1987). Our research supports these perspectives, and we argue that the increased sensitivity to cultural dissonance by male social entrepreneurs is due to perceptions that the group that they belong to (social entrepreneurs) has weaknesses in which their behaviors are perceived as counter stereotypical.

Collectively, our results demonstrate that as cultural dissonance increases, gender becomes a stronger predictor of social entrepreneurial activity. Gender bias and stereotyping has been shown to be a significant factor in the unequal levels in entrepreneurial activity for male and female entrepreneurs cross-culturally (Kelley et al., 2012). Appearing congruent with prevailing stereotypes proves beneficial for entrepreneurship, but appearing incongruent can harm entrepreneurs (Anglin et al., 2022b; Eddleston et al., 2016; Yang et al., 2020). We interpret our findings to propose that female social entrepreneurs may benefit from their lower sensitivity to cultural dissonance, as no action is taken to reduce the discrepancy due to the perceived congruity and low threat. In this way, our research contributes towards a growing stream of research specific to female entrepreneurs and “silver lining effects” (e.g., Greenberg & Mollick, 2017; Johnson et al., 2018; Lee & Huang, 2018), as we find that female social entrepreneurs are less sensitive to these effects of than male social entrepreneurs, and therefore not necessarily at a disadvantage, as suggested by prior studies (e.g., Eddleston et al., 2016; Malmström et al., 2017).

6.4 Limitations

Our work should be considered in light of its limitations. First, some of the pooled data rely on retrospective and self-report questionnaires (GEM), which may introduce individual response bias into the results. We would encourage future scholars to consider replicating the findings using a methodological triangulation approach, which involves using a variety of inductive and deductive tools to enhance generalizability of findings. Second, as noted by other scholars (Hechavarría & Brieger, 2022), we follow precedent and adopt a mission-focused definition of social entrepreneurship (Mair & Marti, 2006). However, considerable variance might exist among social entrepreneurs based on the mission centrality of their organizations, and we echo recent calls for future research exploring how cultural factors influence different types of social ventures (e.g., B-Corps, NGOs, etc.).

Third, our research focuses on one demographic category in considering the effects of cultural dissonance on social entrepreneurship. However, greater nuance may be added due to effects ascribed to intersectionality, which suggests that simultaneous membership in different categories may result in a more complex set of consequences than consideration of any single category (Rosette et al., 2016). Individuals continually draw on a shared understanding of action rooted in cultural knowledge and social categories to justify or make sense of their actions (Swidler, 1986). Our findings suggest that other role congruity theory may be a useful lens to examine how other demographic factors and/or social categories might strengthen or weaken the effects of cultural dissonance on social entrepreneurial action. We would encourage social entrepreneurship scholars to move beyond examination of one social category and its corresponding social roles. Instead, we suggest social entrepreneurs may experience inseparable effects of gender and other categories (e.g., race, class, religion) that could strengthen or weaken the effects of cultural dissonance due to greater complexity in role congruency.

Finally, our research is based on cross-sectional data from a small number of countries (33), which is a general problem in country-level research, although our sample size is consistent with similar studies examining cultural effects (e.g., Stephan et al., 2015). The fact that we found confirmation for our hypotheses, including a robust interaction effect of gender and cultural dissonance, supports the validity of our findings. We suggest that future studies consider other strategies to capture a larger sample size, such as testing alternative nesting structures to group countries with similar cultural practice profiles together.

7 Conclusion

We believe that our research reveals important insights about how cultural values and practices shape social entrepreneurship by the development and application of a unique conceptualization of culture. By operationalizing dissonance between cultural values and cultural practices, we highlight how cultural inconsistencies encourage (or discourage) social entrepreneurial action as a result of action-based discrepancy reduction. We demonstrate the value of a multi-theoretic lens by integrating theories of gender role congruity and action-based models of cognitive dissonance to explain how cultural inconsistencies interact with gender bias and stereotyping to predict social entrepreneurial action.