Entrepreneurs’ Perceived Dissimilarity Promotes HR Formality
and Gender Diversity in Startups
Julia Slegers Student Nr. UvA: 11412879
Student Nr VU: 2606114 16-08-2017 MSc Entrepreneurship Joint Degree Supervisor: Dhr. Dr. Yuval Engel
Statement of originality
This document is written by Julia Slegers who declares to take full responsibility for the contents. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The University of Amsterdam and the VU are responsible solely for the supervision of completion of the work, not for its contents.
Table of Contents
ABSTRACT ... 3
INTRODUCTION ... 4
THEORY AND HYPOTHESES DEVELOPMENT... 8
Gender diversity in startups: An overview ... 8
HR formality and gender diversity ... 10
Linking entrepreneurs’ perceived dissimilarity and HR formality: ... 13
METHODS ... 17
Sample and Procedure... 17
Measures ... 19
Statistical Procedures ... 21
RESULTS ... 22
Limitations and Future research ... 30
CONCLUSION ... 32
REFERENCES ... 33
APPENDIX ... 42
ABSTRACT
Women continue to represent a minority in startups, a situation that might develop into a serious “diversity debt” as these companies scale. Prior studies found that gender diversity in startups can be improved when entrepreneurs adopt more formal human resource (HR) arrangements that can prevent biased hiring. However, it is still unclear which individual characteristics of entrepreneurs may promote HR formality in their startups and contribute to gender diversity among subordinates. Based on the notion of activist choice homophily we propose that just as similarity breeds attraction, dissimilarity breeds activism. Hence, we hypothesize, test, and find support for a sequential mediation model in which entrepreneur’s perceived visible dissimilarity from other entrepreneurs predicts both the time they spend on recruiting and selection new employees and their emphasis regarding formal HR practices. This, in turn, increases the gender diversity in their startup. On the basis of our findings we offer a number of theoretical contributions and point to practical implications as well as future avenues for research on the intersection of HR and entrepreneurship.
INTRODUCTION
Women are severely underrepresented in the world of entrepreneurship (Gompers & Wang, 2017; Minitti; 2009; Thébaud, 2015) and problems associated with insufficient gender diversity may spiral out of control as young startups rapidly scale by hiring large numbers of new employees over a short period of time (DeSantola & Gulati, 2017). This persisting phenomena, wherewith women continue to represent a minority in startups, flies in the face of evidence that gender diversity is not only “associated with increased sales revenue, more customers, and greater relative profit” (Herring, 2009: 208; see also Hoogendoorn, Oosterbeek, & van Praag, 2013) but also that “diversity enriches the work place by broadening employee perspectives, strengthening their teams, and offering greater resources for problem resolution” (Herring, 2009: 208). Indeed, cautionary tales about the implications of accumulating ‘diversity debt’ even made news headlines recently when a series of scandals at Uber, resulting in its founder’s resignation, were attributed to the pervasively toxic masculine culture in the company (Wu, 2017). What then are the mechanisms that explain such detrimental gender imbalance in startups?
Speaking to this question, often while rejecting alternative explanations based on supply-side arguments (e.g., Gompers & Wang, 2017), studies converge around a trickle-down process starting with the historical presence of male majority in entrepreneurship (e.g., Bird & Brush, 2002; Bruin, Brush & Welter, 2007; Calas, Smircich & Bourne, 2009) and persisting through startup-specific, and often biased, human resources management (HRM) (e.g., Baron, Hannan, Hsu, & Kocak, 2007; Cardon and Stevens, 2004). Indeed, men are still far more likely to start new ventures (Bird & Brush, 2002; Bruin et al., 2007; Thébaud, 2015), their initial teams are more likely to contain men than women (Ruef, Aldrich & Howard, 2003; Yang and Aldrich,
2014)1, and these male-led teams are much more likely to receive venture capital, which is the primary engine for further employment growth (Brush, Greene, Balachandra, & Davis, 2017; Verheul & Thurik, 2001). Thus, trying to scale their workforce under extreme uncertainty and lacking public recognition and legitimacy (e.g., Williamson, Cable, & Aldrich, 2002; Cardon & Stevens, 2004), entrepreneurs, the large majority of whom are men, struggle to attract qualified employees outside their existing friends-of-friends’ networks (Ruef et al. 2003). They additionally attempt to differentiate their firms from existing organizations by emphasizing startup-specific employer attributes, such as informal recruitment procedures and flexible work practices (Cardon & Stevens, 2004; Moser, Tumasjan & Welpe, 2017; Tumasjan, Strobel and Welpe, 2011). The consequence of such informal recruitment procedures, where candidates are primarily sourced through entrepreneurs’ own social networks and informal selection procedures are eschewed in favour of subjective fit considerations (Williamson et al., 2002; Ruef et al., 2003), dictate that “those who are demographically dissimilar will experience the greatest obstacles to inclusion and advancement” (Baron et al., 2007: 37). Put differently, in their reluctance to set formal rules and procedures to govern employee recruitment and selection, entrepreneurs may deprive their firms of the utility that HR formality offers as a safeguard against biased hiring (Baron et al., 2007). And because most entrepreneurs are men, this biased hiring is particularly unfavourable for women.
In light of this process, research linking entrepreneurs to gender diversity in startups seem to have confounded the two central elements explaining why women are underrepresented: (1) that men are more likely to start new ventures; and (2) that HRM in these ventures is biased.
1 We note that this excludes spousal or partner co-founding teams where gender heterogeneity is typically higher. Otherwise, entrepreneurial teams tend to be extremely gender homogeneous with men mostly joined by other men and women joining women (e.g., Ruef et al., 2003).
Indeed, it is surprising to find that, despite strong evidence for the relationship between HR formality and gender diversity (e.g., Baron et al., 2007; Yang & Aldrich, 2014), and given entrepreneurs’ discretion over the nature of HR arrangements in their firms (Cardon & Stevens, 2004), the literature still emphasizes ways to increase the share of female entrepreneurs (e.g., Gompers & Wang, 2017; Thébaud, 2015) rather than fixing how entrepreneurs, male and female alike, engage with HRM. More generally, investigations of decision-makers’ characteristics and gender inequality among subordinates have rarely focused on HR formality as a key mechanism, overlooked variation in founders’ beliefs and attitudes as explanatory variables, and hardly ever examined startups as an empirical context (cf. Carnahan & Greenwood, 2017). We therefore pose the following research question:
Whether, how, and to what extent does the relationship between entrepreneurs’ individual characteristics and HR formality shapes gender diversity in startups?
To answer this question, we draw on the notion of Activist Choice Homophily, which predicts that structural barriers shared with a disadvantaged group would promote the desire to help others overcome them (Greenberg & Mollick, 2016). Extending work on homophily in general – the idea that similarity breeds attraction (McPherson, Smith-Lovin, & Cook, 2001) – activist choice homophily goes beyond either induced homophily as based on simple dyadic similarity (e.g., male founders recruit male employees because they have more males in their networks) or interpersonal choice homophily (e.g., male founders recruit male employees because they feel more comfortable with them). Instead, Greenberg & Mollick (2016) find that group-level identification can impact individuals’ decision to support similar others because it entails the recognition of a shared structural barrier (e.g., female or male founders may recruit
female employees because they recognize their disadvantaged position in the startup labour market, sympathize with this position, and have a desire to help). Accordingly, we hypothesize that, controlling for their gender, those entrepreneurs who perceive themselves as dissimilar from other entrepreneurs would be motivated to increase both the time they spend on recruiting new employees and their emphasis on formal HR practices. HR formality, in turn, is expected to positively impact gender diversity through the associated increase in the proportion of female employees.
The current study was designed to make several contributions. First, we have little understanding of why diversity effects occur because research has tended to overlook the possible mechanisms that might underlie such effects (Williams, Parker, & Turner, 2007). While the research of Baron et al. (2007) has started to look into the effects of HR formality as a key mechanism for gender diversity in startups, it did not tell us much about how these HR arrangements come to be and who is responsible for their implementation. This study elaborates on this part by adding an important individual characteristic of founders – their feelings of perceived dissimilarity – as an extension to what we know about HR formality and gender diversity. As a second contribution, we elaborate on the notion of Activist Choice Homophily, as presented by Greenberg and Mollick (2016) who state that it is “the motivation to help someone who shares one’s gender to overcome structural barriers perceived by some to be associated with a shared categorical identity” (p. 25). We therefore broaden the meaning of activist choice homophily and link it to feelings of perceived dissimilarity rather than just objective dissimilarity (e.g., gender). This study points out that, regardless of the entrepreneur’s objective characteristics, his or her perceived dissimilarity may lead to activism and the motivation to help others in need.
To establish these contributions, this study proceeds by providing an overview on the ongoing debate surrounding gender diversity in startups. This is followed by a review of theories about the relationship between HR formality and gender diversity. The theory part ends with an explanatory paragraph on perceived visible dissimilarity and how this term relates to HR formality and gender diversity in startups. We then turn to describing the methods, testing the hypotheses, and reporting the findings. The study ends with a discussion of key contributions and a conclusion which hopes to offer a number practical implications that could be valuable for entrepreneurs, investors, and policy makers.
THEORY AND HYPOTHESES DEVELOPMENT
Gender diversity in startups: An overview
In this section, we will give a small overview of the effects of gender diversity and the current state of gender diversity in startups. First, we have to elaborate on the meaning of a gender diverse workforce. The meaning of the term ‘diversity’ can be open for discussion as it is a broad term which can be used to describe many differences. One form of diversity looks to the so called ‘surface-level’ (Jackson, May, & Whitney, 1995). Surface-level diversity is defined as the differences among individuals in easily identifiable and often directly visible demographic characteristics, which include gender (Williams et al., 2007), “age, sex, and race/ethnicity” (Harrison, Price, Gavin & Florey, 2002, p. 1030). This study will examine a single dimension of surface-level diversity; gender diversity, in the context of the startup workforce.
There have been studies trying to define the effects of gender diversity on teams, e.g. Pearsall, Aleksander and Evans (2008). However, according to Hoogendoorn et al. (2013), most of these studies concerning gender diversity in startups are performed in a lab and lack external
validity. Other studies focussed mainly on the gender diversity in management teams or amongst co-founders (e.g. Ruef et.al. 2003; Welbourne, Cycyota, & Ferrante, 2007).
Yet there are some studies that do focus on the workforce of startups and the effects gender diversity can have. One of these studies is from Hoogendoorn et al. (2013) who came to the conclusion that, “business teams with an equal gender mix perform better than male-dominated teams in terms of sales and profits” (2013, p. 1526). Beyond this direct performance effect, other research points to additional positive effects of gender diversity in terms of, for example, a more diverse pool of knowledge and skills (Hamilton, Nickerson & Owan, 2003). Additionally, Woolley et al. (2010) show that teams with a larger percentage of women perform better because of a higher average level of social sensitivity of the group members. These studies suggest that diversifying the workforce within an organization is beneficial (Hamilton et al., 2003 and Hoogendoorn et al. 2013). Yet relatively few startups actually put emphasis on it (Zimmerman & Brouthers, 2012). When Armstrong (2017) describes the workforce of, in this case, a software startup he says, “they were all white, all young, all childless, mostly male”. Also in Silicon Valley startups, “men dominate the workforce” (Bradshaw & Kwong, 2017). This became a real problem since startups “create essentially all net new jobs” (Kane, 2012, p. 1).
One known explanation for these male-homogeneous teams, is the previously explained homophily principle – the idea that “similarity breeds connection” (McPherson et al., 2001, p. 415). For startups, this means that, because gender is a common basis for homophily (e.g. Reagans, 2005; Greenberg & Mollick, 2016), an already male-dominated startup community tends to remain male-dominated (Kim, 2007). This male-dominance can be traced back to the mid-20th century in which “few women worked outside the homes” (Aldrich & Cliff, 2003, p. 578). Even though this situation has changed drastically and we see a rise of women working
outside the house (Aldrich & Cliff, 2003; Verheul & Thurik, 2001), it is far from equal. For startups the ratio of female employees is 33% out of a sample of 220,000 observations2 (Burton, Dahl & Sorenson, 2016).
Even though we know from previous research that women face contextual barriers, “like social structures, family and organized life”, which influences the access of women into the workforce (Verheul & Thurik, 2001, p.330), the current believes state that prospects for women have never been better. This contributes to an emergence of a ‘gender-blind ideology’, that “claims that gender is no longer important” (Lewis, 2006, p. 458). This gender-blindness, while appearing to be progressive, conceals women’s disadvantaged position and its gendered nature which, in turn, will privilege the masculine (Lewis, 2006). At the same time, Lewis (2006) suggests that “individuals do not have a choice about being recognized as gendered or about which gender they want to be identified with” (p. 463). We agree with Lewis (2006) and argue on previous research that gender diversity and the role for women is still debatable in startups (Burton, Dahl & Sorenson, 2016). We want to contribute to this debate by looking into the underlying effects of the fact that bureaucratic HR structures, tend to promote female participation in the workforce (Baron et al., 2007; Konrad & Linnehan 1995).
HR formality and gender diversity
HR formality is defined as “formal rules, programs, positions, and procedures influencing personnel decision making in an organization” (Konrad & Linnehan, 1995, p. 778). This formality is of great importance regarding the position of women in the workforce as they make the recruitment decisions less subjective (Bielby, 2000). If these decisions are to be made “in absence of formal rules that guide the process”, decision makers “resort to their (often
biased) stereotypes” (Baron et al., 2007, p. 37). These stereotypes in entrepreneurship favour men (Gupta, Turban, & Bhawe, 2008; Gupta, Turban, & Pareek, 2013; Thébaud, 2015), as the “cultural beliefs about gender prescribe lower expectations for women’s competence” (Yang and Aldrich, 2014, p. 306). In contrast, it is shown in previous studies that bureaucracy, formalized and standardized HR practices, in the selection and recruiting of new employees improves the opportunities for women (Baron, et al., 2007, Bielby, 2000). This formal approach is previously mentioned by Baron and Hannan (2002: 12), who argue that if the organization adapts to a ‘bureaucracy’ model it will select individuals “based on their qualifications”. As “bureaucracy is the dominant form of organization in modern society, identifying potential benefits of bureaucracy to organizational women promises a more complete portrait of its gender effects” (DeHart-Davis, 2009, p. 341).
If we now look at the situation regarding bureaucracy and HR formality in startups, we can see a clear difference. Startups are commonly seen as small companies (Decker, Haltiwanger, Jarmin, & Miranda, 2014) where recruiting is found to be quite problematic (Cardon & Stevens, Williamson et al., 2002). This is due to the fact that startups, first of all, are limited in terms of their financial and material resources (Cardon & Stevens, 2004). Next to that, they lack legitimacy as an employer-of-choice (Tumasjan et al., 2011; Williamson, 2000). This, combined with the ad-hoc search of startups for qualified employees and competition with larger and more established employers, results in a high amount of job openings in very short period of time (Nguyen & Bryant, 2004; Stewart & Hoell, 2016). This leads to a limited time frame for a proper recruitment and selection process (Cardon & Stevens, 2004). Because these decisions are made ad hoc and on an infrequent basis, the costs of hiring qualified HR personal are too high for a startup. This results in the resolution of founders and managers to make the decisions
regarding recruitment and selection themselves (Baron et al., 2007; Marlow, 2006). These decision are mostly made without formal rules and procedures (Baron et al., 2007), based on subjective considerations (Ruef et al., 2003; Williamson et al., 2002).
This outcome is not without consequences. When entrepreneurs, mostly males (Ahl, 2006; Kim, 2007), make HR decisions in this informal way, they take away the protection HR formality offers against biased hiring (Baron et al., 2007). As most entrepreneurs are men (Ahl, 2006; Kim, 2007), this biased hiring is causing women to face greater subjectivity as they are not judged purely on their merits (Baron et al., 2007; Yang & Aldrich, 2014). Women are judged based on gender stereotypes in which men are favoured and women face low expectations (Baron et a., 2007). These stereotyped views produce a perceived ‘lack of fit’ that is found to be responsible for the biased judgments about women in work settings (Heilman, 2001). Accordingly, the lack of fit is based on the expectations about how successful or unsuccessful a person will be. These expectations form a strong argument in the HR decisions an entrepreneur has to make (Ibid.).
Concluding, in the gender stereotypes where “men are seen as an advantaged group” (Lewis, 2006, p. 466), women are often depicted as less of a choice, they are excluded or marginalized (Lewis, 2006). As we have seen in the beginning of this paragraph, the formalization of recruitment and selection processes promotes gender equality by looking at the pure merits and qualifications of a candidate (Baron & Hannan, 2002). The lack of exactly that HR formality in startups results in women being excluded. Following this conclusion, we propose the following hypothesis:
Linking entrepreneurs’ perceived dissimilarity and HR formality: the mediating role of
time spent on recruitment and selection
According to the self-categorization theory (Turner et.al., 1987 in Harrison, et al., 2002) people tend to categorize, define and differentiate themselves from others, based on “observable differences in age, race, gender, and the like” (Harrison et al., 2002, p. 1031). Because of this, individuals tend to positively evaluate and identify with persons and groups whose members appear to hold the same observable differences as they do (Harrison et al., 2002). This is supported by the similarity-attraction paradigm, as it states that “people are attracted to, and prefer to be with, similar others because they anticipate their own values, attitudes, and beliefs will be reinforced or upheld” (Harrison et al., 2002, p. 1031). In other words, both self-categorization theory and the similarity-attraction paradigm predict that individuals would hold positive attitudes toward those others who are alike. This is more commonly referred to as homophily, which means that individuals associate themselves with others based on similarity (McPherson et al., 2001). As we see that entrepreneurs tend to recruit new employees similar to themselves (Leung, 2004) we want to elaborate on homophily.
Homophily is further explained in Greenberg and Mollick (2016) who distinguish the term into ‘induced homophily’ and two notions of ‘choice homophily’. Induced homophily is the probability (regardless of conscious choice) that individuals will interact with those with similar characteristics. These characteristics include race, ethnicity, sex and gender (McPherson et al., 2001). In contrast, when looking to the first notion of choice homophily, interpersonal choice homophily is reflected in entrepreneurs who, knowingly, select teams consisting out of those who are similar to them (Greenberg & Mollick, 2016; Ruef et al., 2003). In addition, ‘activist choice homophily’, the second notion of choice homophily, happens when the basis for attraction
are “shared structural barriers stemming from a common group-level social identity and an underlying desire to help overcome them” (Greenberg & Mollick, 2016, p. 2). An example of activist choice homophily is given in the article of Greenberg and Mollick (2016). Female funders of crowdfunding campaigns tend to provide funds to female entrepreneurs because they recognize their disadvantaged position in the startup capital market. They feel “motivation to help someone who shares one’s gender to overcome structural barriers” (Greenberg & Mollick, 2016, p. 25). Put differently, female funders see the other as similar due to a common ground of ‘dissimilarity’ within a certain social context – in this case, their gender.
The term, dissimilarity, refers to the “degree to which an individual and some second entity differ in terms of various characteristics” (Jackson et al., 1995, p. 219). Feelings of dissimilarity can be either objective, also named actual dissimilarity, or subjective, which is also called perceived dissimilarity. Actual dissimilarity refers to an “objective measurement of the degree to which an individual differs from other team members on demographic characteristics and work values” (Hobman, Bordia, & Gallois, 2004, p. 562). Perceived dissimilarity, on the other hand, is a subjective feeling of individuals that tells how different they perceive themselves to be from others (Hobman et al., 2004). We are particularly interested in the perceived dissimilarity of entrepreneurs from other entrepreneurs.
One can be, or feel, dissimilar in many ways, yet here we focus on ‘visible dissimilarity’. Visible dissimilarity refers to feelings of being different based on visible attributes such as age, gender, and ethnicity (Hobman et al., 2004). Note, however that while perceived visible dissimilarity may overlap with actual dissimilarity it is not always the case (e.g., one can think of herself as visibly dissimilar from others even when, more ‘objectively’, outsiders might not agree with that perception). This study examines entrepreneurs’ perceived visible dissimilarity as it
links to activist choice homophily, thus providing possible explanation for the motivation to help those who are also visibly dissimilar within a certain context (i.e., entrepreneurship) (Greenberg & Mollick, 2016; Hornsey, 2008). In a sense, we broaden the notion of activist choice homophily by arguing that just as similarity breeds attraction, dissimilarity breeds activism. By drawing on activist choice homophily it is possible to predict that entrepreneurs who perceive themselves as dissimilar by visible characteristics, from other entrepreneurs, would be more motivated to increase the time they personally spend on recruiting and selecting new employees in the hope of helping others, who might be facing similar structural barriers. The positive effect of motivation on the time spent on recruitment and selection well established in psychology that mention how motivation affects how much time or effort one devotes to a task (Dougherty & Harbison, 2007).
An increase of time spent on recruitment and selection might lead to an increase of HR formality. We argue this as the contrary is proven before, a lack of formal procedures, including HR formality, is associated with shorter amounts of time spent on decisions surrounding selecting and recruiting employees (Baron et al., 2007). We expect that if an entrepreneur spends more time on recruitment and selection, the bureaucracy surrounding these decisions will become more formal. This is arguable as the entrepreneur might feel the urgency to make the procedures for recruitment and selection as efficient as possible. We argue this as we know that entrepreneurs have to pay attention to both the mere survival of the startup as well as to the future (Yadav, Prabhu & Chandy, 2007). When paying more attention, spending more time, to e.g. recruiting a new sales employee, the time that can be devoted to the mere future of the startup drops. As time is a scarce resource (Seshadri & Shapira, 2001), its allocation is a rational decision, hence entrepreneurs might look for solutions to make the recruitment and selection progress more efficient and maybe even delegate it. Especially when a startup gets older, more
full time employees will be employed, which can create “difficulties in organizational coordination and control that prompt” (Baron et al., 2007, p. 60). These findings contributes to the mediating effect of ‘time spent on recruitment and selection’ on HR formality. We therefore propose the second hypothesis:
Hypothesis 2: The positive relationship between entrepreneurs’ perceived dissimilarity and HR formality in a startup is mediated by the time these entrepreneurs spent on employee recruitment and selection.
Combining our two hypotheses, we propose, and test, a sequential-mediation model (see Figure 1) of the role which entrepreneurs perceived visible dissimilarity, the time they spend on employee recruitment and selection, and the HR formality have in shaping gender diversity in their startups. This is captured in Hypothesis 3.
Hypothesis 3: The positive relationship between entrepreneurs’ perceived dissimilarity and startup gender diversity is sequentially mediated by the time they spend on recruitment and selection and HR formality.
METHODS
Sample and Procedure
We conducted a quantitative study amongst entrepreneurs in the Netherlands using an online survey distributed via the Qualtrics platform. A copy of this questionnaire can be found in the Appendix (see appendix A). To recruit participants, we made use of different distribution channels, including networks of accelerators and venture capitalists, social media groups, and a newsletter by the city of Amsterdam. It is unknown how many individuals received our survey link, rendering it impossible to determine the response rate. However, as the recent analysis of Rutherford et al. (2017) shows, “there is virtually no evidence that response rate has any meaningful or consistent influence on relationships in entrepreneurship” (p. 93). We eventually collected a sample of 239 respondents who started the survey. We define an entrepreneur as someone who starts and subsequently manages his or her own business (Blanchflower & Oswald, 1998). To ensure all respondents were entrepreneurs, the first question inquired whether or not the respondent was the founder and manager of the organization. Only respondents with who selected “yes” as their answer were eligible to take part, resulting in 174 remaining respondents.
Since we are interested in entrepreneurs with experience in HR management, another exclusion criterion was pre-set: the respondent’s organization should have at least one fulltime employee, since one cannot engage in HR management without employees. Additionally, we characterized a firm as a startup when it is 10 years or younger (e.g., Baron et al., 2007; Leung, Foo, & Chaturvedi, 2013), excluding all cases older than 10 years from the sample. Finally, some respondents did not complete the questionnaire, resulting in missing values for our key measures. After careful examination, there was no reason to assume the non-response was due to a lack of
clarity or any other reason that might lead to non-response bias. As the number of respondents with an incomplete survey was small, these respondents were excluded from the sample, reducing the sample size to 86 respondents.
We then encountered two outliers in the data that could change our parameters and their means (Aguinis, Gottfredson, & Joo, 2013). Two outlier cases were identified as error outliers as they “are data points that lie at a distance from other data points because they are the result of inaccuracies” (Ibid., p. 282). One of these outliers, a founder of a recruitment company, for example, seemed to have misunderstood our question about his time spent on recruitment and selection for his own firm and instead reported 100%. The other outlier’s response time was very low, five minutes for the entire survey, which indicated to a careless/insufficient effort (Curran, 2016). Next to the response time for this case, we also found that the respondent consistently answered in the same way for almost all items. After taking this into careful consideration, we decided to exclude both cases from the sample.
Therefore the final sample consists out of 84 respondents (60 males or 71%; 24 females or 29%; Mage = 31.43, SD = 9.92). This gender distribution comes close to the population
parameters for entrepreneurs in the Netherlands (KvK, 2017), which consisted in 2016 out of 65% male founders and 35% female founders. Furthermore, the mean age of the firms studied was 3.61 years (SD = 2.34) and the average amount of full time employees was 6.26 (SD = 6.65). Although this study did not target a specific industry, our sample consisted mainly of low-technology companies. We based our coding for industry on the OECD classification (2017). We therefore categorized the industries ‘Computer and Electronics Manufacturing’, ‘Software’, ‘Broadcasting’, ‘Telecommunications’, ‘Information Services and Data Processing’, ‘Other Information Industry’ and ‘Scientific or Technical Services’ as high-technology industries. The
rest was coded in low-technology and respondents who indicated ‘other industry’ or ‘multiple industries’ where coded into the category ‘Other/multiple’ (0 = low-technology, 1 = high technology, 2 = other/multiple). This division resulted in 49% of the startup in our sample operating in low-technology industries, 39% in high-technology industries and the remaining 12% could not be classified and remained with the other/multiple industries category.
Measures
Gender diversity. Gender diversity is measured using Blau’s index for heterogeneity
“which is the most commonly used index of diversity for categorical variables” (Nishii, 2013, p. 1762). Blau’s index of heterogeneity is calculated via 1 − ∑ 𝑝𝑘2 “where p is the proportion of unit members in kth category” (Harrison & Klein, 2007, p. 1211). For gender with the categories being male or female, a result of 0 would correspond to homogeneity and a score of 0.5 would indicate high gender diversity. Respondents had to indicate how many of their employees, females or males, occupied full time, part time, or unpaid (i.e. internship) positions, as well as how many occupied a managerial position. The assumption of normality could not be met for this gender diversity measure. Since our sample is not of considerable size, we used the bootstrapping method to reduce the impact of anomalies (Preacher & Hayes, 2008).
Perceived visible dissimilarity. Perceived visible dissimilarity was measured with two
items from the perceived dissimilarity scale (α=.60) (Hobman et al., 2004). The original measure, using a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree), included 3 sub-scales and 6 items overall (α=.78), addressing visible (2 items), informational (2 items), and value dissimilarity (2 items). As we look for dissimilarity that can explain the increase in motivation and time devoted to HR formality, we will focus on visible dissimilarity as a form of visible characteristics (e.g. sex, age and race). We focus on visible dissimilarity as it links to the
activist choice homophily and can provide answers for the increased motivation to help those who are as visible dissimilar as oneself (Greenberg & Mollick, 2016; Hornsey, 2008). A sample item is “I feel I am visibly dissimilar to other entrepreneurs”.
HR formality. The questionnaire included nine items on HR formality (α=.85).
Respondents were asked to indicate to what extent their startup facilitated each item on a list of formal HR procedures. A sample item is “We provide a formal written policy for an equal opportunity policy”. We retrieved eight of the nine items from Nguyen and Bryant (2004) and one item from Lai, Saridakis, Blackburn & Johnstone (2016). The answers are measured using a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree).
Time spent on recruitment and selection. The time founders spent on recruitment and
selection (M= 14.49, SD=10.56) was measured using a single item; respondents were asked to indicate, using a slider, the percentage (0 – 100%) of the time they spend per week on employee recruitment and selection. This variable did not match the normality assumption. This was mitigated by using bootstrapping (Preacher & Hayes, 2008).
Control variables. On both firm and individual level, we checked for variables that could
affect the hypothesis tests but are not pivotal in this study. Because high technology startups are a difficult barrier for women (Baron et al., 2007; Greenberg & Mollick, 2016; Whitney & Ames, 2004) the firm-level control variable included an industry dummy (low-tech vs. high tech). Additionally, the older a startup gets, the more full time employees are likely to be employed, this can create “difficulties in organizational coordination and control that prompt” (Baron et al., 2007, p. 60) founders to adopt more formal HR to manage this (Baron et al., 2007). Following this reasoning, we aimed at including both the size- and age of the firm as control variables. Unfortunately, as our dependent variable, gender diversity in startups, is also measured using the
proportions of full time employees we cannot include the control variable firm size, in our analyses.
As most entrepreneurs are men (Ahl, 2006), and men are “expected to perceive bureaucracy less favourably than organizational women, because bureaucratic control contradicts notions of cultural masculinity” (DeHart-Davis, 2009, p. 351), we controlled for the gender of the founder. We do this as we want to test if the effects of perceived dissimilarity influence gender diversity. We want this effect to be regardless of the founder’s gender, so we included the gender as an individual control variable (male = 1; female = 0). Additionally, the marital status of male founders is suspected to influence the emphasis founders put on gender equality and is therefore assigned as a second individual control variable (unmarried = 0; married or domestic partnership = 1). Finally, we included, as simple demographic control variable, the age of the founder (Bernerth & Aguinis, 2016).
Statistical Procedures
To test our conceptual model (Figure 1), we used the PROCESS macro for SPSS (version 2.11; see Hayes, 2013). For this study we applied model 6 – a sequential mediation model – as the main statistical procedure to examine the relationships between entrepreneurs’ perceived dissimilarity and gender diversity via the mediating effects of time spent on recruitment and selection and HR formality. The indirect effect was calculated using 5,000 bootstrapped samples for bias-corrected 95% confidence intervals (CI’s)3 (Hayes, 2012).
3
“An indirect effect is considered statistically significant if the CI established (CI at 95%) does not include the value 0” (Igartua & Casanova, 2016, p. 297).
RESULTS
Table 1 provides the means, standard deviations, and correlations for all variables. No significant correlation between perceived visible dissimilarity and gender diversity is observed (r = .07, p = .55). Perceived visible dissimilarity does, however, significantly correlate with HR formality (r = .28, p = .01) and time spent on recruitment and selection (r = .33, p =.002). There is a strong positive and significant correlation between the time spent on recruitment and selection and HR formality (r = .41, p < .001), as well as between time spent on recruitment and gender diversity (r = .29, p = .009). A strong positive and significant correlation between HR formality and gender diversity (r = .45, p < .001) is observed. None of the correlations between the four key variables in our model (variables 8 till 11) have an absolute value greater than .70, minimizing concerns regarding multicollinearity.
Concerning the control variables, we found that, as expected, the founder’s gender correlates negatively with perceived visible dissimilarity (r = -.26, p =.02) and gender diversity (r = -.25, p = .02), indicating that female founders perceive themselves as more dissimilar and that their startups are more likely to be gender diverse. The industry of a startup correlates with gender diversity, positively if it is located in the high technology sector (r = .23, p = .03), and negatively when it is located in the low technology sector (r = -.25, p = .02). The marital status of the founder correlates with the time spent on recruitment and selection (r = .28, p = .01). The age of a firm correlates with the HR formality (r = .29, p = .01). Lastly, the size of a firm (in FTE) correlates with both HR formality (r = .30, p = .01) and gender diversity (r = .33, p = .002).
The hypotheses of this study were tested using the ‘Serial Mediation’ model in PROCESS (Model 6) to estimate regression coefficients and confidence intervals for the all
paths (Table 2). Serial Mediation assumes “a causal chain linking the mediators, with a specified direction of causal flow” (Hayes, 2012, p. 14). Bootstrapped mediation analyses (5,000 resamples) were used to examine the indirect effect of perceived visible dissimilarity on gender diversity via the predicted mediators. An indirect effect is estimated as being significant if zero is not contained within the 95% lower confidence intervals (LLCI) and upper confidence intervals (ULCI) (Preacher & Hayes, 2008). The PROCESS analysis calculates the total and the specific indirect effects of the Independent Variables (IV) on the Dependent Variable (DV). Gender diversity was entered as the DV, perceived visible dissimilarity as the IV, time spent on recruitment and selection and HR formality as the two serial mediators in the following causal order: Perceived visible dissimilarity → Time spend on R&S → HR formality → Gender diversity (see Figure 1). We included six control variables in the analysis, Age Founder (in years), the gender of the founder (male = 1; female = 0), their marital status (married or domestic partnership = 1; single = 0), the firm’s age (in years), high-technology industry (high-tech = 1; other = 0) and low-technology industry (low-tech = 1; other = 0).
Hypothesis 1 stated that HR formality was positively related to gender diversity in startups. As shown in Table 2, the direct effect of HR formality on gender diversity is (b = .10, t(74)= 3.54, p < .001). Since gender diversity is measured with Blau’s Index, ranging from 0 to .5, we can interpret this as an increase of .1 in gender diversity for every 1-point increase in HR formality. Meaning that, for example, if a startup has ten employees and a gender diversity score of .3 (3 females and 7 males), an increase of the HR formality with 1-scale point will change this ratio to .4 (4 females and 6 males).
Table 1. Mean, Standard Deviations, and Correlations
M SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 10 11.
1. Age Founder (In years) 31.43 9.92 - -.08 .52** .42** .27* -.16 -.04 -.04 .02 .1 .07 2. Gender Founder (male = 1; female = 0) .71 .45 - -.03 -.06 -.02 -.17 .19 -.26* -.2 -.19 -.25* 3. Marital Status Founder (married or
domestic partnership = 1; single = 0) .35 .69 - .49** .12 -.15 .02 .06 .28* .17 -.09
4. Firm Age (in years) .28 2.34 - .41** -.08 -.01 .12 .01 .29** .03
5. Firm Size (in FTE) .72 6.65 - -.06 .03 -.11 .01 .3** .33**
6. Firm Industry High tech .49 .50 - .79** .2 .28** .23* .23*
7. Firm Industry Low tech .39 .49 - -.14 -.25* -.21 -.25*
8. Perceived Visible Dissimilarity 2.95 .88 - .33** .28* .07
9. Time Spent on Recruitment and
Selection 14.49 10.56 - .41** .29**
10. HR Formality 2.85 .90 - .45**
11. Gender Diversity
.28 .21 -
Note: N=84; Higher diversity index correspond to higher heterogeneity. *p <.05 (two-tailed test).
Hypothesis 2 suggested that the effect of perceived visible dissimilarity on HR formality is mediated by the time entrepreneurs personally spend on employee recruitment and selection. As shown in Table 2, the direct effect of perceived visible dissimilarity on HR formality was not significant (b = .09, t(75)=.81, p=.42). However, the direct effect of perceived visible dissimilarity on the ‘time spent on recruitment and selection’ was both significant and positive (b = 3.18, t(76)= 2.55, p =.01). This can be seen as an increase of 3.2% in ‘time spent on recruitment and selection’ when the feeling of perceived visible dissimilarity of the founder increases with 1-point on the 5-point Likert scale. This effect shows that if a founder perceives him- or herself as visibly dissimilar from other entrepreneurs, he or she will increase the time they spend on employee recruitment and selection. In addition, the direct effect of ‘time spent on recruitment and selection’ on HR formality was, in line with our expectations, positive and significant (b = .03, t(75)= 3.18, p=.002). This implies that when entrepreneurs spend 1% more of their time on recruitment and selection, this will lead to a .03 increase of HR formality on a 5-point Likert scale. To put this in perspective, a 33.3% increase in time spent on recruitment and selection would be required to increase HR formality in 1-scale point. Based on our results we can say that, in line with our hypothesis, the effect of perceived visible dissimilarity on HR formality is only significant via the mediating role of the time spent on recruitment and selection. Hence, Hypothesis 2 receives support.
Given that Hypotheses 1 and 2 are supported (see Figure 2), we turned to examine the complete sequential mediation model of Hypotheses 3 to determine whether there is convincing evidence to answer our research question. The result (see Table 2) show that the significant total effect of perceived visible dissimilarity on gender diversity (b = -.004, SE = .03, 95% CI [-.06, .05]) was reduced in the direct model (b = -.03, SE = .03, 95% CI [-.9, .02]).
Table 2.
Mediation model (PROCESS, Model 6): Indirect effects of perceived visible dissimilarity (IV) on gender diversity (DV) through time spent on R&S (Mediator Variable 1) and HR formality (Mediator Variable 2). N=84.
DV= Time Spent on DV= HR Formality DV= Gender Diversity Recruitment and Selection Blau’s Index Variables B 95% CI B 95% CI B 95% CI
Effect Boot SE P Boot 95% CI Direct effect of Perceived Visible
Dissimilarity on Gender Diversity -.03 .03 .25 -.08, .02 Effect Boot SE Boot 95% CI Indirect Effect of Perceived Visible
Dissimilarity on Gender Diversity
Total effect .03 .01 [.01, .06] Perceived visible dissimilarity
Time spent on R&S HR formality .01 .01 [.003, .03] Gender diversity
Note. The number of bootstrap samples for the bias-corrected interval is 5,000; IV = independent variable; DV = dependent variable; CI = confidence interval.
As firm size and gender diversity are both measured using FTE, the control variable firm size is not included. Age Founder -.09 -.35, .17 0.002 -.02, .02 .003 -.002, .01 Gender Founder -1.05 -5.79, 3.69 -.10 -.51, .30 -.08 -.17, .02 Marital Status Founder 6.60 2.85, 10.35 -.11 -.45, .23 -.09 -.17, -.01 Firm Age -.83 -1.87, .20 .12 .03, .21 -.001 -.02, .02 High-tech industry 4.59 -2.41, 11.59 .20 -40, .80 -.01 -.15, .13 Low-tech industry -.10 -7.99, 5.99 -.01 -.61, .58 -.05 -.19, .09 Perceived Visible Dissimilarity 3.18 .70, 5.66 .09 -.13, .31 -.03 -.09, .02 Time Spent on .03 .01, .05 .004 -.001, .01 Recruitment and Selection
HR Formality .10 .04, .15 R2 = .28 R2 = .29 R2 = .33
In other words, entrepreneurs’ perceived visible dissimilarity is positively related to gender diversity through the sequential mediating effects of (1) the time spent on recruitment and selection and (2) HR formality (b = .01, BootSE = .01, 95% CI [.002, .03]).
This effect indicates that when the perceived visible dissimilarity increases with 1-point ultimately the gender diversity of the startup increases with 0.01. This can be translated with the previous example, if a startup consists out of ten employees and a gender diversity score of .3 (70%-30% gender split), an increase of the perceived visible dissimilarity with 1-point will lead to a diversity score of .301 (68%-32%).
Figure 2: Unstandardized coefficients for mediation analyses using PROCESS Model 6, 5000 bootstraps (Hayes,
2013). Note. *p < .05, **p <.001
The first side result that we want to mention is the both significant and positive effect of the marital status of the founder to the gender diversity in his or her firm (b = 6.60 SE = 1.88, 95% CI [2.85, 10.35]). We will elaborate on possible arguments for this effect in the discussion. The last side results that grasped our attention was the strong correlation between high tech and low tech industries. We can explain this by looking at the coding of the two variables. As they are coded in exactly the opposite direction, it is understandable that they strongly correlate. The only reason the correlation is not 1 is due to the inclusion of ‘other industries’.
DISCUSSION
In this present study, we examined the sequential-mediated relationship between entrepreneurs’ individual characteristics and gender diversity in startups via the time spent on recruitment and selection and the HR formality of a startup.
Our findings supported the total sequential mediation model by indicating the importance of both the time spent on recruitment and selection and HR formality. With the confirmation of Hypothesis 1 we show the positive effect of HR formality on gender diversity in a startup. With the confirming of Hypothesis 2, both a positive and significant effect was found for the mediating role of the ‘time spent on recruitment and selection’ on the relation between perceived visible dissimilarity and HR formality. Hypothesis 3 suggested a total mediation model indicating that perceived visible dissimilarity does not affect gender diversity without the mediating roles of both the time spent on recruitment and selection and the HR formality. This hypothesis too, was confirmed with our results.
With the confirmation of the hypotheses, this study shows that, regardless of the entrepreneur’s objective characteristics, his or her perceived visible dissimilarity leads to activism and therefore to an increase in the time spent on recruitment and selection. We have shown that just like similarity breeds attraction, dissimilarity breeds activism, as there is both a positive and significant relationship between perceived visible dissimilarity and the time a founder spends on recruitment and selection. Based on its motivational element, found in the time spent on recruitment and selection, we have broadened the notion of activist choice homophily (Greenberg & Mollick, 2016). This notion is also known as the motivation to help someone who shares one’s gender to overcome structural barriers that are associated with some categorical identity (Greenberg & Mollick, 2016). We have linked this notion to feelings of
perceived dissimilarity rather than to solely objective dissimilarity (e.g., gender). In addition, the results indicated a positive effect of the ‘time spent on recruitment and selection’ on HR formality. This might suggest that if a founder spends more time on the recruitment and selection they see the necessity for more formality.
The business case for gender diversity is clear, as within a team, gender diversity, has proven to lead a company to better results (e.g. Herring, 2009). Unfortunately gender diversity itself, is not so easily obtained. Even startups, characterised as non-hierarchical (Cardon & Stevens, 2004), deal with a lack of gender diversity (Burton et al., 2016). Previous findings indicate that HR formality increases gender diversity (Baron et al. 2007). This study confirms these findings but contributes on the literature by moving beyond these results as previous research could not resolve the underlying reason for the positive effect of HR formality on gender diversity.
Subsequently, we elaborated on the research of Baron et al. (2007) who looked into the effects of HR formality but did not supply clear motives for these observed effects. This study has found a motive in the individual characteristic of founders, namely their feelings of perceived visible dissimilarity. We have investigated the feelings of perceived visible dissimilarity of the founder towards other founders (e.g. on objective characteristics like sex, age and race). For their feelings of perceived dissimilarity can explain the occurrence of HR formality. Hence we looked at the impact which perceived visible dissimilarity has on gender diversity within a startup. The results stress that there is indeed a relationship between gender diversity in startups and the perceived visible dissimilarity of their founder. However, at least as important is the conclusion that this is no direct relation. When aiming for an increased gender diversity in startups, two things should be taken into account. First, simply encouraging
minorities, for example women or ethnic minorities, to engage in a startup is not likely to result in an increase in gender diversity if there is no motivation for developing HR formality. Any form of affirmative action to increase gender diversity in startups should therefore be carefully considered and should be accompanied by support in developing formal HR activities. Second, the possibility that HR formality will increase, is much higher when founders of startups perceive themselves as dissimilar, as this drives them to activism, to spend more time on recruitment and selection which, in the longer run, will lead to gender diversity.
In addition to our main finding, our result presented a significant strong and positive correlation between the marital status of the founder and the time they spent on recruitment and selection. To a certain extent, this relates to the literature of Philips (2005), who provided proof for an increased sympathy for gender diversity in the workforce amongst men whom are used to work with female partners. Our present findings and those of Philips (2005) give reason to speculate about a possible relatedness between feelings of sympathy and the marital status of founders. When male founders are married they might feel more sympathy towards women and the barriers they face, these feelings can be translated into the motivation to help women into the workforce, into startups. We did not examine the underlying effects of marital status but propose that future research could explore this correlation between marital status and the time spent on recruitment and selection, therefore the HR formality in a firm.
Limitations and Future research
Like any research, this study has limitations. One is found in the limited sample of entrepreneurs (n=84) we could use for our analyses. This number is due to a low response on the survey and the number of unfinished surveys, which is common when investigating entrepreneurs, as recently has been shown by Rutherford et al. (2017). They reported that founders of startups are
“generally time and cash starved and may simply lack the motivation to complete a survey, particularly in rapidly growing and larger firms” (p.94). However, Rutherford et al. (2017) conclude that there is no evidence that the response rate has influence on tested relationships in entrepreneurship, which empowers the results of our study.
The found results provide an interesting starting point for studying the relationship between perceived dissimilarity and gender diversity in established (and/or larger) companies, especially in those areas where HR formality is less common. One could think of, for example, the assignment of senior management positions and board executives, which are more than once assigned based not solely on skill and experience, but also on a person's network. Our study shows that it is difficult to breach the homogeneous hegemony of (white) males in senior management positions (Joshi, Neely, Emrich, Griffiths, & George, 2015). As a result quota have emerged for senior management (Pande & Ford, 2011) to consist of a certain minimum percentage of females or ethnic minorities, with many positive and negative views on such a policy as a result. The results of this study reveal that future studies might indicate whether developing more and better HR formality for senior management functions could possibly compete with the effectiveness obtained by imposing hard quota. Because objective dissimilarity is not equal to perceived dissimilarity it is not a given that minorities hired based on a hard quota are more likely to increase the gender diversity in the long run, especially when HR formality remains absent. Possible results of additional studies could supply both policy makers and businesses to develop the proper tools to create both gender diverse and effective teams throughout any company.
CONCLUSION
This study makes contributions to theory regarding gender diversity in startups. Our findings demonstrate mediating roles for (1) the time spent on recruitment and selection as well as for (2) HR formality in the relationship between perceived visible dissimilarity and gender diversity. More specifically, we found that perceived visible dissimilarity leads to an increase in the time spent on recruitment and selection. This is due to the notion of activist choice homophily which increases the motivation to help others that face similar and familiar barriers as on self. Secondly we found the positive relation between the time spent on recruitment and selection and the HR formality of a startup. Additionally we found that an increase in HR formality leads to more gender diversity in startups as women are judged on their merits and not on subjective fit. To conclude we want to stress the need for both mediators if one wants to achieve both a positive and significant indirect effect of perceived visible dissimilarity on gender diversity. This is necessary as perceived visible dissimilarity without any mediators, does not influences gender diversity. With that finding, this study presents a solution to alter the gender homogeneous workforces in startups and it gives an answer to the research question central to this study; whether, how, and to what extent does the relationship between entrepreneurs’ individual characteristics and HR formality shapes gender diversity in startups?
REFERENCES
Aguinis, H., Gottfredson, R. K., & Joo, H. (2013). Best-Practice Recommendations for Defining, Identifying, and Handling Outliers. Organizational Research Methods, 16(2), 270-301. Ahl, H. (2006). Why Research on Women Entrepreneurs Needs New Directions.
Entrepreneurship; Theory and Practice, 30(5), 595-621.
Aldrich, H. E., & Cliff, J. E. (2003). The pervasive effects of family on entrepreneurship: toward a family embeddedness perspective. Journal of Business Venturing, 18(5), 573 – 596. Armstrong, R. (2017, June 9). For an Inclusive Culture, Try Working Less. Retrieved from
Hackernoon:https://hackernoon.com/for-inclusive-culture-maybe-less-is-more-87b663662cea
Baron, J. N., & Michael, H. T. (2002). Organizational Blueprints for Success in High-Tech Start-Ups: Lessons from the Stanford Project on Emerging Companies. California Management Review, 44(3), 8-36.
Baron, J. N., Hannan, M. T., Hsu, G., & Kocak, O. (2007). In the company of women. Gender inequality and the logic of bureaucracy in start-up firms. Work and Occupation, 34(1), 35-66.
Bernerth, J. B., & Aguinis, H. (2016). A Critical Review and Best-Practice Recommendations for Control Variable Usage. Personnel Pshychology, 69(1), 229-283.
Bielby, W. T. (2000). Minimizing Workplace Gender and Racial Bias. American Sociological Association, 29(1), 120-129.
Bird, B., & Brush, C. (2002). A Gendered Perspective on Organizational Creation. Entrepreneurship: Theory & Practice, 26(3), 41-65.
Blanchflower, D. G., & Oswald, A. J. (1998). What Makes an Entrepreneur? Journal of Labor Economics, 16(1), 26-60.
Bradshaw, T., & Kwong, R. (2017, April 6). Silicon Valley has a long way to go on gender diversity. Retrieved from Financial times: https://www.ft.com/content/13a73d6e-1951-11e7-a53d-df09f373be87.
Bruin, A. d., Brush, C., & Welter, F. (2007). Advancing a Framework for Coherent Research on Women's Entrepreneurship. Entrepreneurship: Theory & Practice, 31(3), 323-339. Brush, C., Greene, P., Balachandra, L., & Davis, A. (2017). the gender gap in venture
capital-progress, problems and perspectives. Venture Capital, 1-22.
Burton, D. M., Dahl, M. S., & Sorenson, O. (2016). Do startups create good jobs? 1-39.
Calas, M. B., Smircich, L., & Bourne, K. A. (2009). Extending the Boundaries: Reframing 'Entrepreneurship as Social Change' through Feminist Perspectives. Academy of Management Review, 34(3), 552-569.
Cardon, M. S., & Stevens, C. E. (2004). Managing human resources in small organizations: What do we know? Human Resource Management Review, 14(3), 295-323.
Carnahan, S., & Greenwood, B. N. (2017). Managers’ Political Beliefs and Gender Inequality among Subordinates: Does his Ideology Matter more than hers? Administrative Science Quarterly, 1-36.
Curran, P. G. (2016). Methods for the detection of carelessly invalid responses in survey data. Journal of Experimental Social Psychology, 66, 4–19.
Decker, R., Haltiwanger, J., Jarmin, R., & Miranda, J. (2014). The Role of Entrepreneurship in US Job Creation and Economic Dynamism. Journal of Economic Perspectives, 28(3), 3– 24.
DeHart-Davis, L. (2009). Can Bureaucracy Benefit Organizational Women? Administration & Society, 41(3), 340-363.
DeSantola, A., & Gulati, R. (2017). Scaling: Organizing and Growth in Entrepreneurial Ventures. Academy of Management Annals, 11(2), 1-66.
Dougherty, M. R., & Harbison, I. J. (2007). Motivated to Retrieve: How Often Are You Willing to Go Back to the Well When the Well Is Dry? Journal of Experimental Psychology, 33(6), 1108 –1117.
Gompers, P. A., & Wang, S. Q. (2017). Diversity in Innovation. Cambridge: Harvard Business School.
Greenberg, J., & Mollick, E. (2016). Activist Choice Homophily and the Crowdfunding of Female Founders. Administrative Science Quarterly, 62(2), 1-34.
Gupta, V. K., Turban, D. B., & Bhawe, N. M. (2008). The Effect of Gender Stereotype Activation on Entrepreneurial Intentions. Journal of Applied Psychology, 93(5), 1053– 1061.
Gupta, V. K., Turban, D. B., & Pareek, A. (2013). Differences BetweenMen and Women inOpportunity Evaluationas a Function of GenderStereotypes and Stereotype Activation. Entrepreneurship: Theory & Practice, 37(4), 771-788.
Hamilton, B. H., Nickerson, J. A., & Owan, H. (2003). Team Incentives and Worker Heterogeneity: An Empirical Analysis of the Impact of Teams on Productivity and Participation. Journal of Political Economy, 111(3), 465-497.
Harrison, D. A., & Klein, K. J. (2007). What's the Difference? Diversity Constructs as Separation, Variety, or Disparity in Organizations. The Academy of Management Review, 32(4), 1199-1228.
Harrison, D. A., Price, K. H., Gavin, J. H., & Florey, A. T. (2002). Time, Teams, and Task Performance: Changing Effects of Surface- and Deep-Level Diversity on Group Functioning. The Academy of Management Journal, 45(5), 1029-1045.
Hayes, A. F. (2012). PROCESS: A Versatile Computational Tool for Observed Variable Mediation, Moderation, and Conditional Process Modeling. Retrieved from http://www.afhayes.com/.
Hayes, A. F. (2013). Introduction to Mediation, Moderation, and Conditional Process Analysis; A Regression-Based Approach. New York, NY: Guilford Press.
Heilman, M. E. (2001). Description and Prescription: How GenderStereotypes Prevent Women’s Ascent Up theOrganizational Ladder. Journal of Social Issues, 57(4), 657–674.
Herring, C. (2009). Does Diversity Pay?: Race, Gender, and the Business Case for Diversity. American Sociological Review, 74(2), 208-224.
Hobman, E. V., Bordia, P., & Gallois, C. (2004). Perceived Dissimilarity and Work Group Involvement; The Moderating Effects of Group Openness to Diversity. Group & Organization Management, 29(5), 560-587.
Hoogendoorn, S., Oosterbeek, H., & van Praag, M. (2013). The Impact of Gender Diversity on the Performance of Business Teams: Evidence from a Field Experiment. Management Science, 59(7), 1514-1528.
Hornsey, M. J. (2008). Social Identity Theory and Self-categorization Theory: A Historical Review. Social and Personality Psychology Compass, 2(1), 204-222.
Hsu, D. H. (2004). What Do Entrepreneurs Pay for Venture Capital Affiliation? The Journal of Finance, 59(4), 1805-1844.
Igartua, J.-J., & Casanova, J. (2016). Identification With Characters,Elaboration, and Counterarguing in Entertainment-Education Interventions Through Audiovisual Fiction. Journal of Health Communication, 21(3), 293-300.
Jackson, S. E., May, K. E., & Whitney, K. (1995). Understanding the Dynamics of Diversity in Decision-Making teams. San-Francisco: Jossey-Bass.
Joshi, A., Neely, B., Emrich, C., Griffiths, D., & George, G. (2015). Gender Research in AMJ: An Overview of Five Decades of Empirical Research and Calls to Action. Academy of Management Journal, 58(5), 1459-1475.
Kane, T. (2012). The Collapse of Startups in Job Creation. Washington: Hudson Institute.
Kim, G. (2007). The analysis of self-employment levels over the life-cycle. The Quarterly Review of Economics and Finance, 47(3), 397–410.
Konrad, A. M., & Linnehan, F. (1995). Formalized HRM structures: Coordinatig Equal Employment Opportunity or Consealing Organizational Practices. The Academy of Management Journal, 38(3), 787-820.
KvK. (2017, January 16). Jaaroverzicht Ondernemend Nederland. Retrieved from Kamer van Koophandel:
https://www.kvk.nl/download/Jaaroverzicht%20Bedrijfsleven%20Nederland%202016%2 0versie%20US7_tcm109-433766.pdf
Lai, Y., Saridakis, G., Blackburn, R., & Johnstone, S. (2016). Are the HR responses of small firms different from large firms in times of recession? Journal of Business Venturing, 31(1), 113-131.
Leung, A. (2004). Different Ties for Different Needs:Recruitment Practices of Entrepreneurial Firms at Different Developmental Phases. Human Resource Management, 42(4), 303– 320.
Leung, A., Foo, M. D., & Chaturvedi, S. (2013). Imprinting Effects ofFounding Core Teamson HR Values inNew Ventures. Entrepreneurship: Theory & Practice, 37, 87-106.
Lewis, P. (2006). The Quest for Invisibility: Female Entrepreneurs and the Masculine Norm of Entrepreneurship. Gender, Work and Organization, 13(5), 453-469.
Marlow, S. (2006). Human resource management in smaller firms: A contradiction in terms? Human Resource Management Review, 61(4), 467-477.
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a Feather: Homophily in Social Networks. Annual Review of Sociology, 27, 415–444.
Mnitti, M. (2009). Gender Issues in Entrepreneurship. Foundations and Trends, 5(7-8), 497–621. Moser, K., Tumasjan, A., & Welpe, I. M. (2017). Small, but Attractive: Dimensions of new Venture Employer Attractiveness and the Moderation role of Applicants Entrepreneurial Behaviors. Journal of Business Venturing, 32(5), 1-23.
Nguyen, T. V., & Bryant, S. E. (2004). A Study of the Formality of Human Resource Management Practices in Small and Medium-Size Enterprises in Vietnam. International Small Business Journa, 22(6), 595–618.
Nishii, L. H. (2013). The Benefits of Climate for Inclusion for Gender-diverse Groups. Academy of Management Journal, 56(6), 1754–1774.
OECD. (2017, March 22). Glossary:High-tech classification of manufacturing industries . Retrieved from eurostat statistics explained: http://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:High-tech_classification_of_manufacturing_industries
