University of Economics, Prague
Master's Thesis
2020 Pavol Bebjak
University of Economics, Prague Faculty of Business Administration
Masters field: Management
Title of the master's thesis:
Gender-based discrimination in the nonprofit sector. Are nonprofit
organizations hypocritical?
Author: Bc. Pavol Bebjak
Supervisor: doc. Ing. Petr Houdek, Ph.D.
Declaration of Authenticity
I hereby declare that the master's thesis presented herein is my own work, or fully and specifically acknowledged wherever adapted from other sources. This
work has not been published or submitted elsewhere for the requirement of a degree program.
Prague, 25.9.2020 Bebjak
Acknowledgments
I want to express gratitude towards my supervisor, doc. Ing. Petr Houdek, Ph.D., for his guidance and invaluable advice throughout the writing process.
Furthermore, I would like to thank Tomáš Dombrovský from lmc company for
providing data and allowing me to use their job portal for research purposes.
Title of the Master's Thesis:
Gender-based discrimination in the nonprofit sector. Are nonprofit organizations hypocritical?
Abstract:
This thesis explored gender-based discrimination in the Czech Republic. It compares levels of unequal treatment across for-profit and nonprofit sectors using the methodology of a correspondence study. An experimental design with fictitious resumes (N = 170) was implemented to detect gender discrimination at junior-level positions. The analysis of the response rate across both sectors indicated that there is no presence of unequal treatment targeted towards any gender. Contrary to expectations based on some previous studies, nonprofits do not act hypocritically as no discriminatory behavior was detected.
Keywords:
Discrimination, Gender, Nonprofits, CongruenceNázev Diplomové práce:
Diskriminace na základě pohlaví v neziskovém sektoru. Jsou neziskové organizace pokrytecké?
Abstrakt:
Diplomová práce zkoumá diskriminaci na základě pohlaví v České republice. Srovnává úroveň diskriminace mezi neziskovým a ziskovým sektorem pomocí metodologie korespondenční studie. Na odhalení diskriminace na základě pohlaví byl použit experimentální dizajn s fiktivními životopisy (N = 170) které byli rozposlány na juniorní pozice. Analýza odpovědí nedokázala přítomnost diskriminace žádného z pohlaví v obou sektorech. Navzdory očekáváním plynoucím z některých předchozích studí se neukázalo, že by se neziskové organizace chovali pokrytecky, jelikož nebylo odhaleno žádné diskriminační chování.
Klíčová slova:
Diskriminace, Pohlaví, Neziskové organizace, ShodaContents
Introduction 1
1. Theory 4
1.1. Theory of discrimination 4
1.2. Influence of gender in the labor market 4
1.2.1. Difference in preferences 5
1.2.2. Gender-based discrimination 7
1.3. Non-Governmental Organizations 10
1.3.1. Influence of values on organizational behavior 10
1.3.2. Discrimination in NGOs 11
1.3.3. Czech nonprofit job market 13
1.4. Hypotheses 13
2. Methodology 15
2.1. Correspondence studies 15
2.2. Fictitious Resume design 15
2.3. Sample 17
2.4. Response rate measurement 20
2.5. Timeline of data gathering 20
3. Results 22
4. Discussion 26
4.1. Contributions of the study 27
5. Limitations of the study and future research 29
5.1. Limitations 29
5.2. Future research suggestions 30
Conclusion 32
References 35
Annex 42
1
Introduction
Discrimination is defined as an unjust treatment towards members of a particular social group, usually based on prejudices or stereotypes (Becker, 2010). Discrimination often hurts affected social groups as denying them the opportunities in one place, like the labor market, can have further ramifications in other areas of their lives. It can contribute to social exclusion and marginalization of whole groups as discrimination in one area leads to discrimination in another one, thus reinforcing the entire system of unequal treatment in multiple fields like housing, education, or medical care (Lang & Spitzer, 2020). It is proven that discrimination in the access of work or unfair treatment of some group on the labor market can cumulate in overall social disadvantage, manifesting in, for example, lower economic mobility (Blank, 2005). The negative impact of discrimination is the reason why unequal treatment should be studied and why this thesis aims to contribute to its exploration as well. The discrimination of various social groups was analyzed across multiple sectors and fields (Bertrand & Duflo, 2017). This study is asking the question of whether there is gender-based discrimination in the nonprofit sector to advance the knowledge of the subject. The presence of discrimination in the nonprofit sector was not studied extensively, and it begs the question of whether or not nonprofits live by their values when it comes to the treatment of potential applicants.
Discrimination has been studied ever since it was brought up as both economic and social issues. Contrary to other problems of the labor market such as unemployment, discrimination is hard to quantify, and there were and still are numerous attempts to uncover and explain unequal treatment. The fact causes this difficulty in measuring discrimination, that, although it is relatively easy to estimate the economic outcome of a particular social group compared to another, it is much more difficult to quantify, what are the variables that caused this disparity (Drolet, 2002). As such, several methods to study discrimination were developed, including, perhaps, the most widely used one, the correspondence study, which will be used in this thesis as well (M. Bertrand & Duflo, 2017).
Unequal treatment can occur across fields, and employers can discriminate based on the candidate's race, ethnicity, gender, religion, and many other traits and characteristics (Becker, 2010). Such behavior may hurt both the applicant, which can lose some job opportunities and the employer too. Discrimination causes inefficiency as the employer is willing to sacrifice potential revenue from the applicant in order not to be associated with him or her. Furthermore, intentional discrimination is illegal; therefore, employers should do their best to avoid it (Mercat-Bruns et al., 2016).
The disparity between the economic outcome of different social groups participating in the labor market can have many causes. Before attempting to assess what these causes are, however, one has to uncover the presence of the discrimination. This study aims to explore differences (or lack of them) in the treatment of job applicants based on their gender.
2
Although, the discrimination is harmful (Blank, 2005), it is not conditional that it is caused by the animus towards the group that it affects (Phelps, 1972). When it comes to discrimination alone, there are several reasons for its presence, and these can vary across different social groups that are the target of the unequal treatment. Generally, they fall into two categories, either taste- based or statistical discrimination. While some of these can apply to almost all social groups (like in-group favoritism), others are exclusive to specific groups. The statistical discrimination can occur even without a conscious decision being made, as it can be a result of statistical beliefs about a particular group and its outcomes (Phelps, 1972). The existence of statistical discrimination means that unequal treatment cannot be explained away by the presence of racism, sexism, or another sinister undercurrent in society.
Compared to other European countries, discrimination in the Czech labor market has not been studied that extensively and thoroughly and even less so when it comes to discrimination based on gender. Women are one of the groups that experience unequal treatment on the job market the most, which can be showcased on the gender pay gap and many studies that prove that women have a harder time climbing the corporate hierarchy compared to their male counterparts (Yap & Konrad, 2009). Unequal treatment of women in the workplace is one of the reasons (although not the sole one) why the gender pay gap exists (Drolet, 2002). The Czech Republic is one of the countries in the European Union, where the gender pay gap manifests the most having the third-largest gender pay gap in the European Union (European Commision, 2020), and it is one of the reasons why it is worth paying attention to gender-based discrimination. There are multiple possible explanations as to why the gender pay gap exists, including differences between men and women when it comes to employment preferences to the presence of sexism in society (Bensidoun & Trancart, 2018). This thesis, therefore, seeks to contribute to the exploration of gender-based discrimination in the Czech Republic.
Gender-based discrimination is attributed to social and cultural stereotypes that are present in the labor market and the expected roles of the genders (Ridgeway, 1997). This stereotyping has a significant influence on both the choice of individuals when choosing to apply to certain positions and on employers when selecting the best candidate (Burgess & Borgida, 1999). Men are perceived as more fitting for many positions of power by the decision-makers and society in large, resulting in a bias towards women when hiring for positions with more considerable responsibility (Baert et al., 2014).
Moreover, the environment of non-governmental organization and their operations on the market in comparison to regular companies is another area worth studying as NGOs in the Czech Republic grow both in terms of their number and financially, as their earnings rose by 10 percent between years 2011 and 2016 to 189 billion CZK, thus becoming an ever-growing provider of job opportunities (Czech Statistical Bureau, 2019). This thesis also explores whether values that NGOs present externally impact their internal processes, too, as NGOs often signalize their values and their employees are usually attracted by the opportunity to put said values into effect (Sdrali et al., 2016). Combining an exploration of NGOs on the labor market with theories of discrimination should produce new insight into both of these areas. The question of whether and to what extend the gender-based discrimination occurs on the nonprofit
3
as compared to the for-profit sector is, therefore, worth exploring. Furthermore, this study offers insight into whether gender stereotyping has a significant impact on the differential treatment on candidates based on their gender by applying the theory of gender congruence into the research design. The theory of gender congruence states that gender-based discrimination is caused by the incongruence between the gender of the candidate and his or her perceived role in the society as people tend to treat the disparity between one's gender and his or her characteristics unfavorably (Stuhlmacher & Poitras, 2010). The results of the study found no proof for the presence of gender-based discrimination in either of the sectors explored. No significant difference between the treatment of the applicants based on gender across sectors was found either.
This thesis consists of five parts (besides this introduction). In the theoretical part, there is a background laid for the further exploration of the topic of gender-based discrimination and possible explanations for different outcomes of the two genders on the labor market. There is a review of literature in that chapter, too, which offers various explanations for unequal treatment of (primarily) women when it comes to hiring and promotion. Following this part is an overview of the nonprofit sector with a focus on the Czech Republic with implications of the nature of this sector for possible unequal treatment based on gender that can occur within it. Based on the findings mentioned above, two hypotheses to test are formulated.
The second chapter is focused on the methodology of the study with a brief overview of the history of the correspondence studies and further explanation of study design. Following the methodology chapter is the description of the results of the research and the discussion of their possible interpretations. In the end, the conclusion discusses the contribution of the thesis and summarizes the findings. The reader can find the summary of literature used for this thesis together with all the appendices at the end of the thesis
4
1. Theory
1.1. Theory of discrimination
Discrimination in the labor market occurs when the employer does not choose employees based on their qualifications or productive characteristics but the basis of their membership to a particular social group, be it gender, race, ethnicity, religion, age, and others (Guryan &
Charles, 2013). Employers can have a different motivation for this behavior; thus, discrimination can be divided into taste-based and statistical ones.
Taste-based discrimination is caused by the animus of decision-maker towards a specific social group, and as such, an employer who chooses between members of the group towards which he is biased and one towards he is not will always choose the latter option (Becker, 2010). If an individual has a taste for discrimination, according to Becker, he must act as if he was willing to pay something, either directly or in the form of reduced income, to be associated with some persons instead of others (Becker, 2010). Taste based discrimination leads to inefficiency as such an employer will not hire the most productive workers.
Statistical discrimination, on the other hand, occurs because the decision-maker believes that a particular group is less likely to possess characteristics necessary or beneficial for the job. This belief is not caused by animus towards the said group. In effect, statistical discrimination on the labor market takes place when the employer believes that one group will be more productive on average than others and would, therefore, prefer members of that group, applying characteristics of the group to the individual (Phelps, 1972). This belief may be caused by previous experience, wide-held assumptions, and stereotypes about social groups or merely as a result of everyday practice. Beliefs about social groups can be based on accurate or inaccurate data, and as such, statistical discrimination can be further divided into accurate and inaccurate ones (Bohren et al., 2019).
Intentional discrimination is illegal, and governments implement laws to fight it (Mercat-Bruns et al., 2016). Discrimination on the labor market, however, is hard to detect, as it can be subtle, and it is not easy to assess whether the action taken was caused by the intention to behave differently towards a member of a particular group.
1.2. Influence of gender in the labor market
Women are among social groups that face discrimination in the labor market, and as such, discrimination against them has been studied widely. Employers can discriminate against either gender, however, discrimination against women is more prevalent (Baert et al., 2014; et al., 2015; Konrad & Cannings, 1997; Petit, 2007; Yap & Konrad, 2009). Women are discriminated against not only in job markets but also in other markets (for example, credit) (Henderson at
5
al., 2015). Many studies attempted to pinpoint why women are treated unfavorably. These explanations can be divided into two categories according to the cause of the treatment (be it job seeker or employer), supply, and demand-based. Employers can either create unfavorable conditions for women by preferring men (Duguet et al., 2015) or women themselves can prefer jobs that are less paid or valued (Correll, 2004).
There are indicators that men are preferred in the most prominent positions in the job market and better paid than women. The difference in average pay is signified by the gender pay gap that compares average wages of two genders (Drolet, 2002) as well as the fact that there are fewer women on the highest positions of the corporate ledger (Fortune, 2018).
The gender pay gap in the Czech Republic is 20,1 percent (third most significant in the European Union, the average being 14,8 percent) (European Commision, 2020). The industries that have the highest disproportion in average wages of genders are Finance and insurance (39 percent) and Information and communication (31,6 percent). Although the gender pay gap in the Czech Republic is slowly declining (1,5 percentual points from 2010), there are sectors of the economy, where it is stagnant or even growing (European Commision, 2020). As there is no single sufficient explanation of what the causes of this gap are, there are possible explanations ranging from differences in career path (Bensidoun & Trancart, 2018) to actual unequal treatment.
Although the existence of the gender pay gap does not necessarily mean that there is discrimination against women, it is one possible explanation of this disparity. Another reason is that women seek careers and industries that are valued less by the market and prefer to stay out of positions that offer higher pay because of the difference in characteristics between genders and family constraints (Petit, 2007). Women also choose more jobs that are beneficial to society and provides lower wages (Grove et al., 2011).
1.2.1. Difference in preferences
Preference for men in many positions can be caused by differences between men and women (Croson & Gneezy, 2009). Women differ from men in many aspects, and many of them can cause different preferences toward their respective careers. As a group, women are more risk- averse than men (Eckel & Füllbrunn, 2015), which can generate demand for more stable jobs and fewer incentives to leave the current one to search for some better paid alternative.
Women tend to avoid negotiating more than men as they think that society would look upon such behavior unfavorably (Bowles et al., 2007). Women are penalized for negotiating during their job interviews more than men (Croson & Gneezy, 2009). The penalization during negotiations can influence their ability to ask for a higher salary or position or nudge them to the areas that do not require negotiating skills. Women tend to ask less for raise in their salary, which also contributes to the pay gap (Bowles et al., 2007).
6
Women's duties as mothers contribute to the differences in career preferences. Women find themselves in the position where they need to put priority either on their children or their career more often than men (Bertrand, 2013). Women take care of children more than men do;
therefore, more women decide to sacrifice their career advancement to family life (Brockmann et al., 2018). Furthermore, women seem to gain higher satisfaction from having a family rather than having a successful career, which influences their choices (Bertrand, 2013). When deciding whether to advance to a managerial position, women, in general, need to feel that they have higher compensation than men as opportunities to take such jobs usually arise around the same time as a choice to dedicate more time to the family. The more significant incentive to spend time with their families results in women often not feeling high enough motivation for career advancement (Brockmann et al., 2018).
Competitiveness is another area where women differ from men. This fact is observed both in children and adults (Croson & Gneezy, 2009). Differences in competitiveness may have a biological explanation as men have a higher amount of testosterone in their bodies (Colarelli et al., 2006). Testosterone is responsible for aggressive and assertive behavior (Colarelli et al., 2006). On the other hand, there is evidence pointing to the society being responsible for higher competitiveness of men as it seems to be negated in matriarchal societies, where men are the less competitive ones (Croson & Gneezy, 2009). Considering that the Czech Republic is a masculine society (Hofstede Insights, 2020), women should be less competitive in there.
Women are, therefore, not willing to compete for the best jobs on the same level as men are, which naturally results in men occupying these positions.
Women seem to have lower career expectations, which results in not taking on the challenge to get to top positions (Gibson & Lawrence, 2010). The obstacles to reaching the top may be caused by various factors stated above as well as different behavior when it comes to identification with role models. Women have a problem identifying with the highest-ranking employees even when there are enough women in these positions in their company (Gibson &
Lawrence, 2010).
Social and cultural beliefs influence the preferences of genders. People who hold views that members of one gender are inherently better in some tasks will self-assess themselves according to these beliefs. The self-assessing may result in one gender consistently choosing a career path that better reflects these beliefs. Men who believe themselves being better in mathematics are shown to have a preference for jobs that require this skill (Correll, 2004). Furthermore, social stereotypes in what is a traditional career for both genders influence decisions in the area of career advancement (Burgess & Borgida, 1999).
In summary, the differences as mentioned above between genders can result in women having preferences for less paid fields of works and positions as well as lower willingness to negotiate, lower competitiveness, risk-aversion, and pressure to decide between family and career have all influence on the disparity between career choices of the two genders. Additionally, stereotypes and societal beliefs in differences of skills contribute to different preferences too as
7
members of respective genders tend to believe in these stereotypes, which in turn bear influence on their choices (Correll, 2004).
1.2.2. Gender-based discrimination
Gender-based unfavorable treatment on the job market has been studied extensively when it comes to the treatment of applicants on job positions (Baert et al., 2014; Duguet et al., 2015;
Petit, 2007). Papers indicate some level of discrimination of women both in new job applications and advancement inside the company.
In the area of the in-organization job market, two terms have been defined, sticky floors and glass ceiling. Both of them refer to the specific stage of women's careers where it is more difficult for them to advance in the company's hierarchy. Sticky floors phenomenon can be caused by a preference for male managers (Baert et al., 2014) as men are considered better leaders overall, and employees prefer men as their boss (Stoker et al., 2012). Differential treatment does not seem to be confined only to the hiring process but to the promotion process, too, as women have a harder time getting up the corporate hierarchy (Yap & Konrad, 2009).
This apparent obstacle for females to get to the top of the company's ledger is described as a glass ceiling (Saleem et al., 2017). The existence of the glass ceiling can be attributed to the preference for male leaders and social stereotypes as well as differences between genders described in the previous chapter.
Many studies show the employer's preference for male candidates. Based on the theory of taste- based discrimination, one explanation is the sexism of decision-makers (Guryan & Charles, 2013). Recruiters or managers can express misogynic behavior, which would influence their choice of the best candidate (Becker, 2010). This explanation depends on the individual tastes and can offer part of the justification for discrimination occurring on the job market; however, the sexism hypothesis cannot explain all of the unequal treatment as it seems to be a widespread phenomenon, and this explanation would require that sexism is prevalent as well. The Czech Republic, however, was not found as an overly sexist country (Georgetown Institute for Women, Peace and Security, 2020).
Other explanations can be aligned with the theory of statistical discrimination. Motherhood has an impact on an employer's decision on whom to choose as an employee. Potential mothers experience differential treatment when it comes to the selection of jobseekers (Petit, 2007), which can be explained by employer's fear that mothers and young women would prefer to dedicate time and energy to their family rather than focus on their job. In the case of young females, the employer can be concerned about their prospects of leaving their job upon pregnancy (Petit, 2007). Actual mothers in the later stage of their career do not seem to be discriminated against (Petit, 2007).
Social beliefs and stereotypes do not influence only jobseekers and assessment of their skills but also decision-makers during the hiring process. When the position one applies on does not
8
correspond with traditionally perceived roles of the gender, the employer may prefer to choose a candidate that would fit into this position based on the social or cultural stereotypes (Coffman et al., 2017). Discrimination based on stereotypes can be further divided into descriptive and prescriptive ones (Burgess & Borgida, 1999).
Descriptive component of the discrimination can be defined as one that is caused by a woman's nature and is based on the nature of gender as its group attributes, such as productivity, cognitive skills, or psychology (Burgess & Borgida, 1999). Discrimination based on such descriptive component, therefore, results in people naturally attributing gender with some skill or position and in beliefs that members of one gender are more fit for some roles (Stoker et al., 2012). The prescriptive component of the discrimination can be defined as beliefs about attributes, roles, and behaviors to which men and women are expected to conform (Burgess & Borgida, 1999).
Prescriptive stereotype-based discrimination is motivated by the disparity between one's expected and actual social role and results in the hiring of candidates that would fit a given role (male candidate for the male role) (Coffman et al., 2017).
The prescriptive component of discrimination can be further enhanced by the role congruence effect (Konrad & Cannings, 1997). Role congruence comes into effect when people perceive one's current role in the conflict with his or her expected role. According to the theory of role congruence, people dislike when they see such conflict of characters and tend to behave unfavorably towards others with said conflict (Stuhlmacher & Poitras, 2010). Such behavior is reflected when it comes to hiring and promotion, and therefore, women who apply for what is viewed as traditionally masculine roles (and vice versa with men) can face discrimination. This effect takes place also when the applicant's background does not correspond with his or her gender; for example, man, who does housework is viewed unfavorably, however, household work does not affect views on women (Konrad & Cannings, 1997).
Beliefs about the suitability of each gender for a particular position are also impacted by the perceived level of authority, which said the job would bring to an individual (Baert et al., 2014).
In the scenario, when the applicant for employment does fulfill all the requirements for it but hiring him would mean some level of promotion over one's current role, men are preferred. The preference for male candidates in this scenario can be explained by the motherhood penalty as well as the perception of men as with higher potential and, therefore, more eligible for promotion (Baert et al., 2014). Women have, thus, harder time even to start climbing the hierarchical ladder.
Beliefs about differences in productivity of two groups combined with time and energy scarcity can lead to instant dismissal of candidates of a particular social group, as shown in the study by Bartoš et al. (2016). Discrimination does not have to be an act of considerate decision making but merely of simple heuristics as attention is a scarce resource or decision-makers can decide not to pay attention to candidates from a particular social group and thus to discriminate in this way (Bartoš et al., 2016).
9
The organization itself has an impact on the scale of gender-based discrimination that takes place inside of it. Segregation of genders and gender composition in the workplace seems to be one of the contributing factors when it comes to gender-based discrimination inside companies (Kmec, 2008). Lack of interaction between genders causes held beliefs about another gender to reinforce and drives both descriptive and prescriptive stereotyping, which results in discrimination. Although descriptive stereotypes do not seem to go away, even in companies with a higher rate of interaction between genders, the level of discrimination is decreased (Ridgeway, 1997).
Another factor when it comes to gender discrimination is the sector of operation of the company. On the one hand, areas that are traditionally occupied by women can be penetrated by men relatively easily (Kmec, 2008). On the other hand, when it comes to male-dominated fields, women have a harder time getting a position in such areas (Ridgeway, 1997). One of the explanations as to why this happens can be found in social and cultural stereotypes of what are traditionally women's professions. Women are believed to perform worse on gender-neutral tasks, and therefore, employers can prefer male applicants in general over female ones, especially when the job is not traditionally viewed as female one (Ridgeway, 1997).
If the company (or whole sector) is dominated by one gender, in-group bias takes place. The dominance of the industry by one gender means that members of one gender would prefer the candidate of the same gender over another one. The percentage of females inside the department that is hiring candidates is influential too, as if too small, women do not have enough power over decisions. The probability of employing a woman seems to be highest when women account for around a third of the department as upon reaching the approximate parity in gender representation, and the department will no longer consider gender a decisive factor (Yoder et al., 1989).
In sum, the reasons for differential treatment based on gender in the labor market can be explained in three different ways. The first one is the taste-based discrimination caused by the sexism of decision-makers, who simply prefer candidates of one gender because of their animus towards the other one. The second reason for discrimination can be found in the theory of social stereotypes and role congruence theory. According to these theories, discrimination occurs as a result of beliefs about what typical positions and roles for each gender might be. Decision- makers would prefer candidates that fit into stereotypical roles about gender and would be biased against ones that defy traditional roles. As a result, some sectors of the economy, as well as some companies can become dominated by one gender, and in the case of women, it causes this sector to prefer male employees. The preference for one gender may be caused by the third reason for differential treatment, which is in-group favoritism, which contributes to the choice of one gender over another, especially in companies that are dominated by said gender.
10
1.3. Non-Governmental Organizations
Non-governmental organizations (NGOs) are defined as organizations that do not seek to create profit to redistribute. On the contrary, they try to utilize their funds for other goals (Neziskovky.cz, 2019). These goals can be various, usually aiming for the benefit of society.
NGOs, however, do not form a homogenous group as they vary in size, management style, sources of income, and what their main activity is. Their core business can be an aid to a particular group of people, management of specific activities (sports clubs), educational, extra- curricular, or religious events. The main activity of the NGO will probably bear influence on its choices in managing it. Furthermore, nonprofits have different stakeholders compared to their for-profit counterparts as their primary income is not predicated on the satisfaction of their customers but rather on their donors or members who pay membership fees.
The management and organizational structure of NGOs, in general, gets closer to that of the corporate company as NGOs strife to be more efficient (Dees & Anderson, 2003). The drive for efficiency implies that the processes that govern NGOs and profit-seeking organizations are somehow similar. The structure of NGOs larger in size is similar to the corporate structure, too, as NGOs have their CEOs, executive boards, middle and lower management (Dees & Anderson, 2003). While the structure may be alike, the people who work in NGOs are different from those in their corporate counterparts (Sdrali et al., 2016). People working in NGOs are more concerned with social issues and have less interest in money (Sdrali et al., 2016). They are also motivated by different incentives, such as helping others or contributing to the better world.
Differences in motivation imply that despite being similar in structure, NGOs can operate differently from corporations as people who work there may have different beliefs about how to manage an organization.
1.3.1. Influence of values on organizational behavior
Both companies and NGOs have values that they tend to communicate both externally and internally (Murphy & Davey, 2002). Organizations highlight that they are equal opportunity providers and that they do not discriminate under any circumstances. These values, however, are usually not reflected in the organizational behavior itself as they tend to influence external communication and treatment of customers and partners more than they do employees (Murphy
& Davey, 2002). Employees themselves can prefer to act upon established social values or unofficial structures of power rather than values of the organization they work in (Foote, 2001).
Organizations try to showcase their non-discriminatory policies by various statements or awards for being an equal opportunity employer. As shown by Marques (2010), however, many companies put up these statements as a response to lawsuits concerning their discriminatory behavior.
The communication of one's values is not confined to the corporate environment as other organizations such as NGOs or schools signalize that they treat everyone equally as well. This signalization, however, does not always correspond with reality, and actions taken to showcase the values of the organization are often superfluous. The superfluity can be demonstrated by
11
the example provided by Bradley et al. (2018). They showed that Chief Diversity Officers hired by faculties had little impact on increasing actual representation of minorities on school campuses as they did not implement policies that would reduce or eliminate barriers that prevented higher levels of diversity on the universities in the first place (Bradley et al., 2018).
Furthermore, values themselves are often not included even inside the strategies of NGOs or charities as they use them more as a tool for presentation and communication with the external environment (Foote, 2001).
While organizations tend to apply their values when it comes to dealing with groups of external stakeholders, they do not seem to accent them as much in managerial strategies of the organization itself (Foote, 2001). The disparity is caused by the relative hardship to successfully embed organizational values into the behavior of the employees as they prefer to act based on social or cultural norms and even to follow unofficial structures of power rather than obey a company’s values (Murphy & Davey, 2002). This inconsistency applies to NGOs, too, as pointed out by Foote (2001) and results in NGOs being sensitive in dealing with their clients in a fair and just way while overlooking the same behavior when it comes to their employees. The discrepancy between values and acts is not limited to interactions with various stakeholders but can also be present in the handling of finances as larger nonprofits with more funds were more likely to show irregularities in their financial statements (McDonald & Goodman, 2020).
This disparate application of NGOs' values may, therefore, prevent the values to prohibit the unequal treatment when it comes to choosing new employees as decision-makers may not consider framing the choice of the best candidate in terms of the values but rather in terms of social norms. It is reasonable to expect that the employees of the NGOs are also prone to the same biases and stereotypes as employees elsewhere.
1.3.2. Discrimination in NGOs
From the similarities and differences between NGOs and companies, drawn above, it is not evident whether discrimination would work in the same way in these two types of organizations.
On the one hand, similarities in organizational structure and processes (including hiring ones) may point in the direction of similarities in unequal treatment too. On the other hand, differences in the type of employees that are attracted to work in respective types of organizations may create different outcomes when it comes to discrimination.
The discrimination against workers that belong to a particular social group inside of NGOs has not been studied extensively, and as such, there is no definitive answer as to whether unequal treatment is a widespread phenomenon compared to corporations. There are many NGOs that fight for equality of opportunities on the labor market, and discrimination would, therefore, directly contradict their values. Organizational values, however, do not seem to be a significant factor when it comes to their influence on internal processes (Murphy & Davey, 2002).
As for gender discrimination inside NGOs, there is no clear evidence pointing for or against it.
There are, however, some indicators that NGOs are exposed to the same influences that gender has on a labor market. Women seem to be more attracted to the NGO sector as it offers most of
12
the jobs in the field that women prefer, like administration or finance (Zieleńska, 2012). NGOs also seem to have more understanding of maternity leave, thus offering more flexible contracts, and their jobs often require less traditionally masculine skills like competitiveness or risk- tasking (Zieleńska, 2012). For these reasons, women dominate entry-level positions in the NGOs, and men tend to adapt to this more feminine environment, not vice versa (Faulk et al., 2013).
On the other hand, when it comes to higher-ranking positions inside organizational leadership, there is a disparity in the representation of men and women in the third sector overall as male to female ratio in the management positions of NGOs is favorable towards men (Third Sector, 2017). The women hold 40 percent of leadership positions inside nonprofits, while they represent 60 percent of overall NGO employees in Poland (Zieleńska, 2012). Similarly, women hold 47 percent of jobs in the organization's leadership in the nonprofit sector while representing 65 percent of positions in NGOs in the UK (Third Sector, 2017). The disproportion between men and women inside the top posts even inside nonprofit organizations can be attributed to either demand or supply reasons, as explained in the chapters above. It can serve as a proof that nonprofits are no exceptions when it comes to their employees behaving based on social and other stereotypes.
Another indicator of disparity between the two genders in the nonprofit sector is the existence of the gender pay gap (Faulk et al., 2013). Although it is smaller than in other industries, this difference seems to be caused by men settling for more modest wages rather than women earning more, as they accept the industry standard (Faulk et al., 2013). The measurement of the gender pay gap in the Czech Republic indicates that the sectors, in which NGOs are mostly represented (e.g., Human health and social work activities or Education) have a higher level of the gender pay gap compared to the economy overall (25,6 and 23,5 percent respectively compared to country average of 20,1 percent) (European Commision, 2020). This finding runs contrary to results from abroad (specifically the USA), where the higher presence of the nonprofits decreased the gender pay gap in measured sectors (Faulk et al., 2013).
The underrepresentation of men on the lower levels of the NGO structure can be attributed mainly to the higher interest of women in the third sector, but gender stereotypes can contribute to this phenomenon too. According to the ACEVO study that monitors the third sector in the UK, men often feel discriminated against when on lower positions inside NGOs, and they think that women are preferred when applying for these positions (CharityJob, 2018). Preference for women can be explained by the prevalence of women in the decision making positions inside HR departments and in-group favoritism as well as by the role congruence theory, as low- ranking jobs inside the third sector can be considered more feminine (Konrad & Cannings, 1997).
Based on the findings mentioned above, it is reasonable to expect that there would be no significant gender-based discrimination during the selection process in NGOs compared to one in the for-profit companies. Conversely, it can be expected to see preferences for women if male candidates would be seen as engaging in feminine activities. This expectation can be drawn from the fact, that disparity between the gender of the candidate and his characteristics
13
would cause the recruitment employees to see the applicant as not fitting the societal norms and as such undesirable (Burgess & Borgida, 1999). Employees in the nonprofits are expected to fall in large for the same biases and prejudices like everyone else. NGOs also have similar processes to companies when it comes to hiring, the process itself should, therefore not cause the difference.
1.3.3. Czech nonprofit job market
The number of NGOs registered in the Czech Republic has been rising steadily during the last few years to 151 364 in the year 2019, and thus, it can be deduced that the number of jobs created by these organizations has increased as well (Czech Bureau of Statistics, 2020).
The data from job portal jobs.cz can support the growing importance of nonprofits. Job portal jobs.cz allows NGOs to advertise their job opportunities for a lower price than usual, and therefore it keeps track of how many positions NGOs offer on the job market. In the year 2019, the number of NGOs that advertised their vacancies on jobs.cz and its affiliate job portals was 570. These NGOs announced 6 422 vacancies (counting only those advertised for at least 14 days and excluding short term vacancies), making it 11 positions per NGO on average (lmc, 2020).
The attractiveness of nonprofit jobs is similar to its level in the for-profit market. In support of this statement is the fact that the average number of responses for the job advertisement of the NGO is 12,3 compared to 16,5 in the for-profit market according to the job portal jobs.cz. While NGOs are still not as attractive employees, the average number of responses is significantly higher than that of recruitment agencies with an average response rate of 8,1 (lmc, 2020). The average number of responses in the nonprofit sector might look surprising as the positions in the nonprofit industry are generally less paid. Lower salary can be, however, compensated by higher flexibility as the percentage of flexible and part-time jobs is significantly higher in the nonprofit job market (Zieleńska, 2012). The difference in the job opportunities between the nonprofit and for-profit sector also comes in the type of positions NGOs tend to advertise the most. These are usually connected to social work or nursing (more than 70% of all the positions) as they are related to the primary activities these organizations engage in (lmc, 2020).
1.4. Hypotheses
Considering information gathered on gender discrimination on the job market, one can conclude that female applicants are expected to perform slightly worse when looking for a job when applying to the vacancy in the company compared to men. The gender-based disparity can be demonstrated by various previous similar studies (Baert et al., 2014; Duguet et al., 2015; Petit, 2007). Women are considered worse candidates in general for various factors such as social stereotypes, or mere sexism, especially when applying to what is perceived as traditionally male roles.
It can be further concluded that, when it comes to NGOs, unequal treatment of female applicants may be in large part negated by the fact that the nonprofit sector has a higher percentage of
14
female employees, and therefore in-group favoritism can take place when choosing between candidates for the position. As employees of NGOs are also susceptible to the same biases as everyone else, if the particular candidate would display perceived congruity between his or her gender and his or her social role, they would prefer a different candidate.
Hypothesis 1: There will be no significant preferential treatment for either gender inside NGOs compared to corporations.
Hypothesis 2: There will be discrimination of candidates with perceived role incongruity (male candidates engaging in feminine activities).
15
2. Methodology
2.1. Correspondence studies
This study used the correspondence methodology to uncover the existence and the scope or gender discrimination inside the hiring decisions of NGOs. Correspondence studies are used extensively to study the phenomenon of discrimination in labor markets and elsewhere as it offers high control over variables that are thought to be responsible for the unequal treatment (M. Bertrand & Duflo, 2017). Compared to different methods like audit, which consists of pairing real people to pose as applicants, correspondence study offers a relatively easy way to get in some decision from the employer (Bertrand & Duflo, 2017).
Although the correspondence study has become a staple in the field of studying unequal treatment, it is not without its flaws. Heckman (1993) points out that fictitious resumes sent to the employers can have unobserved characteristics that cannot be accounted for in the resume itself (such as the particular name that can have a more positive connotation to him/her) although the decision-maker can base his/her decision on these variables. This criticism was targeted primarily on studies focused on ethnic and racial discrimination as they were much more widespread at the time; Baert (2015) points out that this critique is not relevant when it comes to studies on gender discrimination. He found out that the variance in unobserved determinants that cause the bias is comparable for both men and women (Baert, 2015).
Another point of criticism towards correspondence studies is that it does not reveal possible discrimination happening during the later stages of the recruitment process and, therefore, is confined only to a narrow decision-making process. Said an only small group also decides for employees who come into contact with the resume and who, as showcased in the study of Bartoš et al. (2016), are susceptible to biases and often act heuristically. Furthermore, correspondence studies are unable to detect discrimination inside the organizations because vacancies are primarily filled by internal candidates, and there are indications that unequal treatment is happening there as well (Yap & Konrad, 2009). Therefore, correspondence studies are leaving a big part of the labor market unexplored (M. Bertrand & Duflo, 2017). Lastly, the ethical concern about such studies can be raised as its methodology requires to react to job offers as a candidate that will never show up in later stages of the recruitment process and thus depleting the recruitment staff of their valuable time and by extension the company or organization of its resources that could have been allocated elsewhere. Despite the validity of the bottlenecks mentioned above, the correspondence study methodology presents a standard in the field and thus will be used in this study too.
2.2. Fictitious Resume design
Resumes of the fictitious applicants were created based on the template provided on job portal jobs.cz to create the illusion of the average job seeker. The name of the job seeker was picked randomly based on name frequency data from the top 10 most frequent names and surnames
16
used in the Czech Republic based on the website kdejsme.cz (2020). The name was indicative of the applicant's gender. Email address and home address differed as well. However, the city was the same across all the resumes to eliminate the possible preference for people from a town (if the application targeted different city, reason for it was stated in the cover letter, expressing interest to move to the city of the vacancy for personal reasons). The skills and other qualifications of the applicants were constructed based on the research of the job market to fit the most frequent vacancies opened in the nonprofit sector and did not differ across three fictitious resumes.
To be able to uncover possible unequal treatment based on gender, resumes were made in three different copies, differentiated by their gender and gender-based interests to create one gender congruent resume. The theory of gender congruence says that gender-based discrimination comes into place when there is a conflict between what is expected of the given gender and how the given candidate fulfills these expectations. While Resumes A and B were sent to both companies and NGOs to get a comparison of unequal treatment across the sectors, resume C was sent only to vacancies in the nonprofit sector to uncover possible bias when treating gender incongruent candidates there. Table 1. presents the structure of the resume and the variables that were changed across them.
Table 1. Structure of the resumes
Resume A Resume B Resume C
Same across all the resumes
Education
Years of experience Working experience Certificates
Language skills PC and other technical skills
Education
Years of experience Working experience Certificates
Language skills PC and other technical skills
Education
Years of experience Working experience Certificates
Language skills PC and other technical skills Points of
difference
Female gender
indicated by the name Female interests Email address Home address
Male gender indicated by the name
Male interests Email address Home address
Male gender indicated by the name
Female interests Email address Home address
The incongruence of the Resume design C was achieved by indicating that the individual (represented by said resume) engages in activities that run contrary to what one would expect within the resume of a particular gender. In the case of this research, the Resume C was manipulated to show feminine interests (volunteering in the organizations focused on gender equality and interest in gender and female studies) for the resume with the male name. The other two resumes were designed to express an interest in the activities associated with various
17
NGOs (e.g., educational and environmental activities) To indicate a similar amount of interest in the nonprofit sector.
As several resumes can be sent to one employer, the education and working experience varied slightly to ensure that resumes were not be detected as part of the experiment. These variations, however, were only in the name of the school and employer as the job description and type of studies was the same across all the resumes. As schools and employers differ in quality, the high schools assigned to resumes were of the same kind and relatively same quality, and Universities, if required, were the same. As for the working experience, companies or organizations with smaller sizes were assigned as it is more unlikely that recruiters would know these, and thus, they would not have any preference.
Resumes were sent out during five months or until the sufficient sample of job advertisements and reactions was gathered. During this period, alerts on job portals were set up to notify about any new posting of positions that would satisfy the conditions of the study design.
2.3. Sample
The correspondence study targeted both nonprofit and for-profit organizations to uncover whether the difference between the discrimination in these two sectors exists. Companies were chosen from industries that are gender-neutral (meaning that they do not have more than 60%
employees of one gender), according to Czech Statistical Bureau (2020), to account for possible in-group favoritism. This decision was made to exclude sectors that are overrepresented by one gender. As the size of the organization does not seem to affect the rate of discrimination within it, all sizes of the organization were subjects of the study (Baert, 2018) and the size of the organization was recorded together with other characteristics of the organization that might have an impact on the rate of discrimination within it. These included the location, sector, and whether the organization is Czech or international. Positions were chosen from job advertising portal jobs.cz and its affiliate portal, prace.cz, as it provides a platform for the NGOs to advertise their positions for advantageous price conditions. These positions can, therefore, be found in relative abundance there compared to other job portals, as of August 2020, there were 560 positions advertised in the nonprofit section of the job portal (jobs.cz, 2020) compared to mere six offers advertised on the webpage neziskovy.cz or 53 postings on Czech website of job portal indeed.com.
Moreover, jobs.cz portal offers a quick and easy way to react to job offers and keep track of the timeline of communication and time of reaction. It also served as an archive for all the responses to the responses on job vacancies. The permission to use the portal jobs.cz was acquired beforehand from the employees of the company lmc s.r.o. That manages portal jobs.cz together with its affiliated portals.
In order not to overwhelm the recruitment employees, resumes were sent to a maximum of two positions advertised by one organization. The limitation of two resumes per organization also
18
ensured that the same recruiter would not be a decision-maker for more than two applications, and the dataset would not be influenced too heavily by a single recruitment employee.
Furthermore, the position with fewer requirements that could be assessed more quickly by the recruitment staff were prioritized. These requirements included years of experience, technical and language skills, need to provide cover letters and references from previous employers.
Conditions mentioned above should contribute to the relative quickness of assessment from the part of the recruitment worker evaluating the resume.
Vacancies that were picked were either entry or junior level positions with a maximum of two years of experience required. Furthermore, junior-level jobs have similar requirements across the market, and therefore, they can be reacted to by the same resume. Thirdly, as pointed out by Baert et al. (2014), the level of seniority and responsibility has a significant impact on the rate of gender-based discrimination. Keeping all targeted vacancies on the entry or junior level allowed to exclude this possible bias and removed one variable that would influence the results from the mix.
Resumes were sent to the job advertisements for positions in both nonprofit and for-profit sectors that require similar qualifications in terms of education, knowledge of languages, and years of working experience. By doing so, the samples from both areas were comparable, and the response rate could be compared. Due to relative scarcity of the positions that would fit precisely into all of the requirements for both sectors concerned, the methodology of three judges was used to determine which jobs will be finally chosen as similar enough for the study.
This method was used, for example, in the study by McKinney and Davis (2003) and consists of an evaluation of the similarity from three independent judges, which should know the job market.
The most frequently advertised positions in the nonprofit sector, according to data from lmc s.r.o. are shown in Table 2. The CVs were designed to fit these positions to find the highest quantity of suitable vacancies. As vacancies in the area, administrative work was relatively plentiful both in for-profit and nonprofit sectors; they were picked as primary recipients of the reactions, and some jobs that required both administrative and social work were added for the nonprofit sector to have enough resumes for comparison.
19
Table 2. The most frequently advertised positions in the nonprofit sector
Position Number of advertisements in
2019
Percentage of total
Social worker 1525 46
Nurse 543 16
Medical assistant 390 12
Administrative worker 296 9
Instructor 229 7
Psychologist 181 5
Teacher 180 5
Source: lmc, 2020
Positions were chosen based on the following criteria:
● Maximum up to two years of experience required
● Required language and technical skills to fit resume designs
● Categorization as either Administrative worker (further divided into Administrative clerk, Salesclerk, and Accountant) or Social worker
● No need to provide references from previous employers
● Possibility to react directly from job portal jobs.cz (employer did not ask to contact him directly)
Throughout the data gathering period, 170 reactions to vacancies were sent to 72 companies and 74 NGOs through the job advertising platform jobs.cz. The number of resumes submitted for each resume type is indicated in Table 2., followed by Table 3., which presents the distribution of reactions across pre-defined professions.
Table 3. Resumes are sent by resume design.
Resume design Nonprofit sector For-profit sector
Female resume 30 40
Male resume 30 40
Male resume with implied gender incongruence
30 -
20 Table 4. Resumes are sent by advertised positions.
Position Female resume Male resume Male resume with
implied gender incongruence
Accountant 19 15 9
Administrative clerk 27 34 7
Social worker 14 12 12
Salesclerk 10 9 2
2.4. Response rate measurement
The response rate was determined as a percentage of responses from overall reactions on job postings. Only responses asking for a personal meeting, phone call, or another form of expressing interest from the organization posting the job advertisement were considered when counting the response rate. Conversely, a negative response was recorded when the recruiter rejected the application or did not respond at all. If the communication from the organization occurred after the expression of the interest, the reaction time was marked. To make the process simple for the responding organization, potential invitation to further stages of the recruitment process was politely declined with an excuse of no more interest.
When reacting to the job offers using fictitious resumes, the only email address was provided as a possible channel of communication, making the organization respond to the sent resume through email. Email communication ensured that the experimental design was harder to detect for the recruitment employee and made communication archivable. After each resume sent, the time, name of the position and organization and its characteristics stated above (size, sector, and location) as well as a CV sent were marked down for further analysis. If there was a response from the organization, its time and nature were marked as well. If the organization did not respond in two weeks, the response was marked as "no answer."
The response rate was correlated with individual resume design to determine whether there are differences in it for each given CV. The analysis was done in both the nonprofit and for-profit sectors to uncover any possible discrepancies in the reasonable discrimination rate in these two sectors.
2.5. Timeline of data gathering
The data were gathered in four months, from the 5th of March to the 20th of July 2020. The relatively long duration of data gathering was caused in part by the epidemic of Covid-19 and ensuing quarantine, during which vacancies posted on job portals, that would fit the research criteria were relatively scarce.
21
The scarcity of job advertisements was more prevalent in the nonprofit sector compared to for- profit one, resulting in the completion of the research regarding companies one and a half months earlier, in May 2020, while research regarding NGOs carried on to July 2020. This disparity in the phases of data gathering was accounted for by using the variable of collection phase (0,1) in the regression analysis of the results.
Table 5. presents variables that were tracked and used for analysis in the next chapter.
Table 5. Variables used
Variable Values
Resume design Female, Male congruent, Male incongruent
Response type Positive response (1), Negative response, including no response (0)
Organization type NGO, Company
Response time Number of days
Field of employment Finance, Administration, Sales, Social work Field of NGO
Care for socially disadvantaged individuals Work with minorities and foreigners
Education and cultural activities
22
3. Results
Binary regression analysis of results was undertaken to uncover significance of differences between samples divided by particular variables. The regression models used in the analyses are described below:
logit (reaction type) = β0 + β1Xi1(resume design) + ℇi
logit (reaction type) = β0 + β1Xi1(organization type) + β2Xi2(resume design) + ℇi
logit (reaction type) = β0 + β1Xi1(organization type) + β2Xi2(resume design) + β3X31(phase) + ℇi
logit (reaction type) = β0 + β1Xi1(NGO field) + β2Xi2(resume design) + ℇi
logit (reaction type within 3 business days) = β0 + β1Xi1(NGO field) + β2Xi2(resume design) + ℇi
logit (reaction type) = β0 + β1Xi1(organization type) + ℇi
Table 6. presents the results of regression analyses for various factors that were taken into consideration.
Table 6. Regression analyses
Analysis Variables Result
Discrimination based on gender Resume design χ2 = 1,93
df = 2 p = 0,35 Discrimination based on gender
across sectors
Resume design Organization type
χ2 = 1,07 df = 3 p = 0,78 Discrimination across sectors
influenced by phases of data gathering
Resume design Organization type Phase
χ2 = 2,91 df = 4 p = 0,57 Effects of gender congruence Resume design (limited to gender
congruent and gender incongruent male)
χ2 = 0,13 df = 2 p = 0,72 Fields within NGOs Resume design
NGO field
χ2 = 2,61 df = 4 p = 0,62 Attention discrimination Resume design
Organization type
χ2 = 0,68 df = 3 p = 0,88 Comparison of response rate
across sectors
Organization type χ2 = 0,02
df = 1 p = 0,87 Results of binary logistic regression of discrimination across sectors indicate, that there is no significant connection between gender (male or female) and response type (positive and
23
negative) (χ2 = 1,93; df = 2; p = 0,35). Similarly, regression analysis after adding organization type variable (nonprofit and for-profit) shows that response type has no significant association with organization type either (χ2 = 1,07; df = 3; p = 0,78).
As data gathering in the nonprofit sector proceeded one and half months beyond the gathering in the for-profit one, dummy variable (first and the second phase of the gathering) was added into the dataset to control for the difference in timelines. Binary logistic regression shows no significant association between phases of the data gathering and response type (χ2 = 2,91; df = 4; p = 0,57).
The effect of gender congruence was tested by comparing the response type for gender incongruent resume design with the other two designs. As mentioned above, binary logistic regression shows that there is no significant link between resume type and response (χ2 = 0,87;
df = 2; p = 0,28). When comparing only samples of male resumes with and without gender congruence, results do not indicate any statistically significant link either (χ2 = 0,13; df = 2; p
= 0,72).
After dividing job postings into four fields of employment (administration, finance, social work, and sales), a non-parametric Wilcoxon Signed-Rank Test was performed. Results for every area are presented in Table 1. There was no significant difference in the treatment of the applicants in any of the fields that were part of the study.
Table 7. Results of the Wilcoxon Signed-Rank Test for each field of employment
Field of employment Result
Administration p = 0,23; statistic = 16,5a
Finance p = 1; statistic = 5a
Sales p = 0,35; statistic = 0a
Social work p = 0,78; statistic = 12a
Nonprofit organizations were further divided into three groups:
- Care for socially disadvantaged individuals - Work with minorities and foreigners - Education and cultural activities
Connections between these groups, resume design and response type were tested by binary logistic regression test. Results show no significant link between the aforementioned variables (χ2 = 2,61; df = 4; p = 0,62).
To check for possible attention discrimination (refusing resumes of undesirable candidates without further consideration), a binary regression test was performed on the dataset consisting of responses that happened within three business days (counting for a possible delay in emails) in the opening from the date of the job application. This test did not show any significant link
24
between response type and resume design within the pool of aforementioned interactions and organization type (χ2 = 0,68; df = 3; p = 0,88).
Charts 1. and 2. present the overall results of the correspondence study in both the nonprofit and for-profit sectors. As described above, no significant discriminatory effect was found in either of the industries.
Chart 1. Response rate based on gender in the nonprofit sector
Table 8. Response rate based on gender in the nonprofit sector
Positive response Negative response No response Female 23,3 percent 36,7 percent 40 percent Congruent
male
16,7 percent 46,7 percent 36,6 percent Incongruent
male
13,3 percent 43,3 percent 43,4 percent
SD 5,1 percent 5,1 percent 3,3 percent
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Female Congruent Male Incongruent Male
Positive response Negative response No response
25
Chart 2. Response rate based on gender in the for-profit sector
Table 8. Response rate based on gender in the for-profit sector
Positive response Negative response No response
Female 20 percent 40 percent 40 percent
Male 17,5 percent 40 percent 42,5 percent
SD 1,8 percent 0 percent 1,8 percent
The binary logistic regression was performed to see whether there was any significant difference in response rate across sectors. The difference in response rate was proven to be statistically insignificant by this regression (χ2 = 0,02; df = 1; p = 0,87). Table 9. presents the overall comparison of results across sectors.
Table 9. Cross-sector comparison
For-profit sector Nonprofit sector
Positive response rate 18,8 % 20 %
Negative response rate 33,3 % 40 %
Average response time 5,9 days 8,8 days
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Female Male
Positive respose Negative response No response
