Token Status and Management Aspirations Among Male and Female Employees in Public Sector Workplaces

Abstract

This study brings the literature on tokenism together with multiple theoretical lenses on the formation and social construction of men’s and women’s career aspirations. The study builds on comprehensive survey data from the Danish public sector. Results show that, after controlling for alternative explanations with respect to both personal life situation and differences between occupations, token status has a significantly negative effect on women’s management aspirations, while it has no effect on men’s aspirations to management. Furthermore, these findings are generalizable across occupational contexts. At the same time, however, analysis across occupations show that token women are mainly to be found in occupations where women have relatively high managerial aspirations. Token women are therefore characterized by aspirations to management positions, but their status as tokens minimizes these aspirations.

Keywords

career aspirations token gender studies public management

Introduction

The extensive vertical gender segregation in the labor market exclude women from the top of the hierarchical structures and limits the resource pool from which society can pick future managers (Charles, 2003; Hultin, 2003; Reskin, 1993; Reskin, McBrier, & Kmec, 1999; Tomaskovic-Devey et al., 2006). The leaky pipeline theory describes the progressive attrition of women between job entry and managing career positions (Mariani, 2008). Despite government efforts and interventions to promote gender equality (e.g., employment equity, pay equity, nondiscrimination acts, etc.), women continue to be underrepresented in management positions. As Schweitzer, Ng, Lyons, and Kuron (2011) point, equal opportunities for advancement might formally reduce gender inequality, but if male and female employees do not use the opportunities equally because of different levels of aspirations, it does not solve matters of recruitment. To better explain the gendered career gap, Schweitzer et al. (2011) suggest we consider the career aspirations of those within the pipeline. In this article, we focus on gender differences in career aspirations with a particular focus on the correlation between gender token status and management aspirations. Hence, we explore the association of vertical gender segregation in the labor market with another feature of the labor market, namely, horizontal gender segregation. Because of persistent—both vertical and horizontal—gender segregation of the labor market, researchers from different social sciences have brought considerable attention to exploring the experiences of norm-breaking individuals in gender-atypical jobs and occupations (Cognard-Black, 2004; Floge & Merrill, 1986; Kanter, 1977a, 1977b; Williams, 1992; Yoder, 1994; Zimmer, 1988). Employees, who are minority representatives of their gender in the workplace (e.g., female firefighters), are known as “tokens” (Kanter, 1977a), and the interactional patterns and social processes specific to tokens are known as tokenism.

Often building on case studies utilizing qualitative methods, gender tokenism research finds that token women typically experience isolation, stereotyping, and other difficulties that restrain their integration in the workplace and even more their ascension up the hierarchies (Floge & Merrill, 1986; Greed, 2000; Kanter, 1977a, 1977b; Turco, 2010). Conversely, token men often appear to benefit from the token position (Fairhurst & Snavely, 1983a; Williams, 1992, 1995; Yoder & Sinnett, 1985). Referring metaphorically to the possibilities for hierarchical ascension, it has been stated that female tokens hit a “glass ceiling” (Bendl & Schmidt, 2010; Yoder, 1991) or stick to a “sticky floor” (Yap & Konrad, 2009), whereas token men as, for example, male nursery teachers, nurses, or midwifes ride a “glass escalator” (Williams, 1992). In contrast to case studies exploring work-related experiences among gender tokens, quantitative studies on tokenism focus on the effects of token status on outcomes like wages, promotions, and turnover (Budig, 2002; Cognard-Black, 2004; Hultin, 2003; Maume, 1999). The quantitative studies show mixed findings regarding effects of gender token status. Hence, there seems to be a need for studies that focus on gender token status effects not just across men and women but also across different occupational contexts.

This quantitative—cross-sectional—study brings the literature on tokenism together with recent theories and research on the formation and social construction of men’s and women’s career aspirations in gendered organizational work contexts (Cohen & Swim, 1995; Correll, 2001, 2004). We explore the following questions:

Does workplace token status affect management aspirations of women and men?

Does workplace token status affect men’s and women’s management aspirations alike?

Are the effects of workplace token status to men and women generalizable across different occupational contexts?

The study analyzes unique survey data from workplaces within a wide range of public sector occupations in Denmark (13 in all, see list of occupations in Table 1). The data consists of an equal number of women and men from each of the 13 occupations. This sample composition allows us to compare workplace tokens with nontokens and women with men across a range of highly different occupational settings, each representing different opportunity structures with regard to, for example, career ladders, occupational responsibilities, and salary.¹ Further controlling for working hours, tenure, and family situation, our study has good prospects for determining whether workplace token status has implications that are generalizable across different gender and occupations. Hence, empirically, our study expands on existing token research by exploring the possible associations between gender token status and men’s and women’s career aspirations, with a particular view to the generalizability of gender token processes across a range of occupational settings. Theoretically, our contribution lies in the joining of theories on tokenism with theories on the formation of career aspirations in gendered work contexts.

Table 1.

List of Occupations Included in Data and Response Rate.

Occupation	N	Number of male and female respondents
Social and health workers	213	M = 103, F = 110
Teachers in primary schools	226	M = 106, F = 120
Physicians	180	M = 92, F = 88
Health care professionals	251	M = 117, F = 134
Office and IT staff	250	M = 107, F = 143
Academic staff in public administration	249	M = 120, F = 129
Professional caretakers in 24-hr care institutions for vulnerable children and youths	229	M = 109, F = 120
Professional caretakers in day care institutions	211	M = 90, F = 121
Technical staff and cleaning	180	M = 87, F = 93
Teachers in youth educations	336	M = 172, F = 164
Researchers	177	M = 71, F = 106
Police and prison staff	154	M = 75, F = 79
Employees in the armed forces	162	M = 61, F = 101
Sum	2,818	M = 1,310, F = 1,508

Research on Workplace Tokenism

Token Experiences

In her seminal study of Fortune 500 Company “Indsco,” Kanter (1977a, 1977b) finds that token female sales managers shared numerous experiences related to their token status. The token women at “Indsco” felt they had to work harder (than their male peers) to receive recognition for individual achievements, and any mistakes were closely scrutinized and criticized. Highly visible female tokens felt pressured to prove their professional worth as compared with their male peers. Kanter concluded that the token “Indsco” women were restrained and did not enjoy the same career opportunities as men.

Kanter’s observations and theoretical analyses laid the ground for many later studies of tokenism in organizations, and her results have been tested and reproduced among token women in a number of organizational settings: female police officers (Gustafson, 2008; Ott, 1989), firefighters (Yoder & McDonald, 1998), engineers (McIlwee & Robinson, 1992), construction workers (Greed, 2000), Wall Street professionals (Roth, 2004), and female physicians (Floge & Merrill, 1986). Many of these studies confirm that women in gender-atypical occupations encounter the negative experiences described by Kanter. Some studies also present results contradicting Kanter’s perspective, however, most notably Hammond and Mahoney’s (1983) study of seemingly well-placed and confident female Appalachian coal miners.

In her original interpretation, Kanter theorized that the token processes were universal and ascribable solely to the numerical relationships between social categories in the workplaces. In Kanter’s definition, token persons belong to the 15% minority in skewed groups, while the 85% majority, according to Kanter, make up the “dominants.” Later tokenism researchers have defined tokens differently. Some have emphasized the importance of being a single representative for ones’ group (Konrad, Kramer, & Erkut, 2008), while others have utilized more relaxed statistical measures (Budig, 2002).

Kanter (1977a) assumes that “the same pressures and processes can occur around people of any social category who find themselves few of their kind among others of a different social type” (p. 240). According to this understanding, all token persons (females and males, Whites and Blacks, etc.) should expect to encounter the same hardships found among female tokens at “Indsco.”

Gender Differences in Tokenism

Kanter’s expectation to find a universal or purely numerical token effect has not been met. Studies of male tokens have often found that male tokens appear to benefit from the token position. Male tokens may experience high visibility and to some degree even contrasting role encapsulation. With men, however, these processes often seem to bring positive outcomes, including career advantages, responsibilities, or wages (Fairhurst & Snavely, 1983b; Floge & Merrill, 1986; Williams, 1992, 1995; Yoder, 1991; Yoder & Sinnett, 1985; Zimmer, 1988). Various elements of positive male tokenism have been identified in a number of different settings: among male nurses (Floge & Merrill, 1986; Heikes, 1991; Williams, 1992), elementary school teachers (Cognard-Black, 2004; Williams, 1992), librarians and social workers (Williams, 1992), and flight attendants (Young & James, 2001). But diverging results also exist. For example, one study found that male social workers seemingly ran against “another ceiling,” experiencing antimale bias in promotion (Atwater & Van Fleet, 1997, p. 603).

One of the general conclusions from the male token studies is that Kanter’s numerical token theory provides an inadequate understanding of the token processes. If token processes were attributable to numerical relationships between social categories alone, then token status for men and women should result in similar experiences and similar social consequences. Understanding the inequalities stemming from tokenism therefore requires looking “beyond the numbers” (Yoder, 1991).

Explanations Based on Status and Culture

The most widely accepted explanation for the differences between the experiences of male and female tokens relates to differences in the ascribed status between the sexes. In sociological and psychological research, it is widely recognized that gender serves as a principal marker of external status (DiTomaso, Post, & Parks-Yancy, 2007; Ridgeway, 1991; Ridgeway, Johnson, & Diekema, 1994; Sachdev & Bourhis, 1987). In culturally inscribed images, gender is associated with notions of competence, value, and worth, and hence affects subordination and superordination in groups (Berger, Ridgeway, & Zelditch, 2002; Driskell, 1982; Driskell, Olmstead, & Salas, 1993; Rashotte & Webster, 2005; Ridgeway, 2001). Gender is commonly described as a diffuse status characteristic, and women are generally ascribed a lower social status than men. These status differences may affect interactional patterns in organizations in important ways and lead to different experiences for male and female tokens. According to the status perspective of tokenism, persons belonging to lower status categories tend to suffer the negative token experiences described by Kanter and “hit the ceiling,” whereas those belonging to higher status categories benefit from the minority position and “ride the escalator” (McDonald, Toussaint, & Schweiger, 2004; Roth, 2004; Yoder, 1994, 2002).

In one recent study, Turco (2010) finds that the token experiences of White women and African American men in the leveraged buyout (LBO) industry differ substantially. Token women suffer from token experiences, while token Black men do not. This result represents an important supplement to the status perspective on tokenism, because both women and African American men are generally ascribed a lower status than White men. Turco explains the differences by supplementing the status perspective with a cultural perspective on tokenism. According to Turco (2010), the women in her study face a clash between cultural images of the ideal worker specific to the LBO industry and widespread societal notions of femininity. African American men are also atypical in the industry and represent a category with lower status than White men, but in the LBO industry, they are not afflicted by the same cultural incompatibility that afflicts women. In the end, Turco suggests that tokenism processes and outcomes could be more locally determined and more variable than hitherto acknowledged. The status and cultural perspectives on gender tokenism—outlined above—most notably diverge in their expectations of generalizability across occupational contexts. The status perspective expects different token experiences between the sexes, but no differences between different occupations. The cultural perspective, in contrast, expects different patterns in token processes between occupations to the extent that cultural images and ideals are specific to the occupational contexts.

Career Aspirations and Tokenism

As reviewed above, the relatively wide-ranging token research typically based on case studies finds that gender token status is associated with certain experiential patterns: stereotyping and role encapsulation. These patterns seem to have negative bearings for some (women) and positive for others (men). Recent research on the formation and social construction of career aspirations indicates that gender tokenism processes could affect career aspirations among employees. In the following, we will develop this theoretical association between gender token status and employees’ career aspirations.

In two companion articles presenting evidence from experimental as well as “real world” (i.e., nonlaboratory) studies, Correll (2001, 2004) finds associations between cultural beliefs about gender, men’s and women’s individual self-assessments, and the formation of career aspirations. Particularly interesting for the present study is the fact that Correll finds a tendency among women to downgrade self-assessments and to dampen career aspirations in situations where gender is highlighted as a relevant negative status characteristic.

Correll’s analyses support theories on Stereotype threat (Spencer & Steele, 1999; Steele & Aronson, 1995; Steele, Spencer, & Aronson, 2002). Stereotype threat is the expectation that one will be judged on the basis of social identity group membership rather than actual performance and potential (Roberson & Kulik, 2007). Research finds that stereotype threat creates anxiety and lowers the performance among persons that are afflicted (Roberson & Kulik, 2007).

Belonging to gender token or gender minority groups in an organization increases the salience of negative stereotypes and promotes greater stereotyping (Ely, 1995). Furthermore, token or minority position is associated with stronger perceptions of stereotype threat (Roberson & Kulik, 2007; Roberson, Deitch, Brief, & Block, 2003). Token or minority position in an organization, therefore, may invoke stereotype threat and resulting performance loss.

Individuals facing stereotype threat may choose different coping strategies (Block, Koch, Liberman, Merriweather, & Roberson, 2011). However, it is not unlikely that women experiencing stereotype threat may react by suppressing or not developing management aspirations. In our societies, manager positions are often regarded as a male gender–typed position, and empirical studies have shown that many people ascribe stereotypical manager traits to men more often than to women (Heilman, 2012; Roberson & Kulik, 2007). When gender-based stereotype threat lowers women’s performance and expectations for performance in an organization, women’s management aspirations will likely suffer. An organizational environment with pronounced gendered stereotype threats, therefore, may suppress management aspirations among women.

Career and management aspirations seem to be particularly likely victims of the hypothesized association because of the highly prevalent gender typing of managerial skills as male. The hypothesized association between token status and career aspirations seems to be consistent with the findings of Maume (1999), Hultin (2003), and Cognard-Black (2004), that token status is associated with career outcomes.

Furthermore, research points out the association between token status and not only negative experiences but also negative expectations toward future work situations. In experimental studies, Cohen and Swim (1995) and McDonald et al. (2004) find that women who anticipate being tokens are more likely than nontoken women to have negative expectations toward future work situation. This result is consistent with Schweitzer et al.’s (2011) findings that the gender gap in precareer salary expectations is greater in traditionally male-dominated fields.

King, Hebl, George, and Matusik (2010) find that, for women, token status is associated with “a psychological climate of gender inequity” and further that a climate of gender inequity is negatively associated with a number of indicators for a psychological working environment, among others work commitment. This association between token status, climate of gender inequality, and commitment appears to underpin our hypothesis concerning the association between token status and career aspirations.

Common to the studies above is that they indicate associations between gender token status and features regarding the individual employee’s work orientations, including career aspirations, and/or career associated notions like expectations and commitment. The basic mechanism is that token status increases the salience of a person’s gender and evokes processes of stereotyping and stereotype threat, which can affect self-assessment, performance, and career aspirations among token persons.

Male Advantage as Usual?

Even though we find that there are good reasons to assume that gender token status can affect men’s and women’s career aspirations, research on gender token status often suffers from methodological weaknesses. Much research builds on single-sex samples, that is, samples consisting entirely of token women or token men and often drawn from a single occupation (Fairhurst & Snavely, 1983a; Hammond & Mahoney, 1983; Heikes, 1991; King et al., 2010; McDonald et al., 2004; Yoder, 1985). Other studies comprise both sexes in gender-skewed occupational settings, but still study only one or a few occupations (Floge & Merrill, 1986; Gans, 1987; Gidengil & Vengroff, 1997; Gustafson, 2008; Roth, 2004; Spangler, Gordon, & Pipkin, 1978; Turco, 2010; Young & James, 2001). Typically, the first type of study conveys important insights into the experiences, attitudes, and behavior among token persons. However, by only exploring the experiences of the gender token persons, this kind of study has a weak methodological foundation for relating the experiences and attitudes observed with token status or gender, because no intercategorical comparisons are available. The second type of study provides a better foundation for differentiating between tokens and nontokens but still faces difficulties establishing that gender token experiences are due to token status and not gender (or ascribed societal status). As Budig (2002) notes, if you study a sample of only token women (or token men) or if you are unable to compare experiences from token persons with those of nontokens, you will be unable to ascertain whether the experiences, attitudes, and responses you observe specifically relate to the token status of the persons explored or to other characteristics. Furthermore, if exploring tokenism in a single occupational setting, you will not be able to establish the generality of category-related token experiences across different occupations. In the long run, of course, an array of single-occupation studies can pile up evidence and together suggest more general patterns.

Few large-scale quantitative studies of tokenism outcomes exist. Budig (2002) finds no support for the hypothesis that male tokens enjoy special privileges compared with men generally. She, therefore, concludes that token research based on case studies often confuses “male advantage as usual,” with a token-related male advantage.

In the present study of the possible association between gender tokenism and management aspirations, “male advantage as usual” could relate to basic gendered career patterns with respect to management aspirations. Mainiero and Sullivan (2006) find that, typically, among professionals, men follow an alpha career pattern while women follow a beta career pattern. The alpha career pattern is “linear and career-centric” (Mainiero & Sullivan, 2006, p. 134). The beta career pattern “represents the career of someone who values family and makes adjustments to career for the sake of balance” (Mainiero & Sullivan, 2006, p. 147). The difference between these career patterns could be associated with gender, but not with gender token position. In contrast to Budig, Hultin (2003), Maume (1999), and Cognard-Black (2004), all find some support for the notion of a “glass escalator” for male tokens. According to these studies, token male employees have better prospects for internal promotion than do equally qualified women (Hultin, 2003), and better chances of moving to a management position (Cognard-Black, 2004; Maume, 1999). Ng and Wiesner (2007) find that men compared with women are more likely to be hired when they are minority candidates and less qualified. Furthermore, employment equity directives tend to benefit men but not women. Their findings suggest that men appear to enjoy a structural (as usual) advantage and the “glass escalator effect.”

Hypotheses

Based on the different positions in the token literature above we pose the following—contradicting—hypotheses.

In accordance with Kanter’s theory, we hypothesize the following:

Hypothesis 1 (H1): Token status is negatively correlated with management aspirations among all employees regardless of gender and occupation.

However, the status perspective on gender tokenism predicts that female and male tokens are affected negatively and positively, respectively. Accordingly, we pose the following hypothesis:

Hypothesis 2 (H2): Token status is negatively correlated with management aspirations among women, but positively among men.

As status differentials between the two genders seem to comprise all or most societal contexts, the status perspective seems to indicate that the consequences of tokenism are independent of occupational specificities. Supplementing the status perspective, Turco’s theory on the cultural basis for tokenism predicts that the outcome of tokenism will vary between occupational contexts with different cultural images and ideals. We will discuss our results with a view to the possible interaction between tokenism and occupational context. Variations in the importance of tokenism across occupational contexts will support the cultural perspective, while uniform importance (or nonimportance) will run counter to it.

Budig’s (2002) critique of existing token research concludes with a conception of “male advantage as usual” (p. 260), which is consistent with Williams/Ackers work on the gendered organization (Acker, 1990; Williams, 1995) and the conception of typical gendered alpha and beta career patterns (Mainiero & Sullivan, 2006). According to these conceptions, token status does not contribute on its own to workplace-related inequality processes. Accordingly, based on Budig, Williams/Acker, and Mainiero and Sullivan, we pose the following hypothesis:

Hypothesis 3 (H3): Management aspirations are higher among men than among women, but not correlated with token status.

Research Design, Data, and Measures

To test the hypotheses outlined above, we use a large, unique data set collected in 2009-2010 among Danish public employees (Madsen, Holt, Jonassen, & Schademan, 2010). The questionnaire was sent to 8,759 public employees with a response rate of 56%.² The survey was designed as a gender-stratified random sample of employees and managers within 13 different occupations (see Table 1). Many public sector jobs in Denmark are female dominated (e.g., health care professionals, professional care takers in day care institutions, and social and health workers) and others are male dominated (e.g., police and prison staff and employees in the armed forces). Hence, in attempt to get a fairly equal amount of male and female respondents within each occupation, our sample of respondents oversample the nondominating gender of each occupation. Data were also collected among a 14th category of public employees—or rather among a residual category of public employees, namely those who were not covered by the 13 occupations. For reasons of clarity when interpreting the data, we chose to exclude the 214 respondents in the 14th job category from the analysis. Furthermore, 1,873 already hold a management position and are therefore also excluded from the analysis.

Comparing the respondents with the nonrespondents of the survey showed no significant differences. The response rates for males and females are almost equal (48% for males, 52% for females). Fewer from police and prison staff and technical staff and cleaning have answered (respectively, 47% and 48%) than, for example, academic staff in public administration (66%) and secondary school teachers (63%). However, as we use occupation as control variables and hence do “within profession”–studies, these response rate differences do not invalidate our statistical findings. In the statistical analysis, professional caretakers in 24-hr care institutions for vulnerable children and youths are used as reference category, as this occupation is the one within which employees score their management aspirations closest to average.

Even though Denmark has higher levels of gender equality than most countries when it comes to workforce participation, the job market is fairly gender-divided (Bloksgaard, 2011; Emerek & Holt, 2008). Women are disproportionately employed in caretaking jobs in the public sector. At the same time, Danish men and women are—again compared with other countries—fairly even when it comes to taking care of family and home, their levels of education, and level of activity in spare time. Hence, because of the relative equality and norm for both sexes to have gainful employment, one might argue that the Danish case is a critical case, making it a conservative test of our hypotheses.

The design and data of the article are, however, not without caveats. The cross-sectional data allow us to analyze the association between token status and management aspirations of women and men across highly different occupational settings each representing different structures of opportunity regarding career ladders, job responsibilities, and so on, and hence through occupation dummies, control for these differences in occupational settings. However, cross-sectional data are susceptible to potential omitted variable bias and reverse causation bias. The first concern is that employees with token status on certain other variables (unobserved and hence not controlled for by the researcher) differ from employees with non–token status, and that these differences may be correlated with employee’s management aspirations. Hence, we cannot be completely certain that token status in itself is causing differences in management aspirations or whether the observed association is driven by something affecting both token status and employee management aspirations. Hence, caution in drawing causal inferences is advisable. However, the fact that an estimate could be biased does not mean that it is. Furthermore, the fact that the data are Danish, and hence collected in a relatively homogeneous setting, diminishes the possibility of omitted variables in regard to other individual characteristics such as ethnicity, race, and religion that might cause token status on other personal characteristics than gender and hence affect management aspirations. Furthermore, studying across occupational settings allows us to indirectly control for level of education, as the studied occupations require different levels of education (e.g., to become a researcher in Denmark, you need to have a PhD degree, and to be a school teacher, you have to have a 4-year bachelor degree in teaching from a certified university college). Finally, we control for a variety of personal life situation variables.

However, a second concern with cross-sectional data is that observational data often imply that we do not know the causal order among variables. We know that employee gender is first in causal order compared with token status and management aspirations but we cannot be certain that management aspirations are second to token status and not vice-versa. This is, however, a matter we discuss in our interpretations of findings.

Measures

Dependent variables

The dependent variable in the analysis is management aspiration. To obtain a comprehensive measure of this, we used two items: one asking “To what degree would you like to become manager at your present workplace?” the other asking “To what degree would you like to become manager at another workplace?” Both questions were to be answered on a scale of 1 to 5, the endpoints indicating “not at all” and “to a very large degree.” The answers to the questions were added together into an additive index (Cronbach’s α = .73) measuring “interest in management position” (M = 1.92, SD = 1.03). The rationale for combining the two items is that such measure is in line with the theoretical argument of token effects, namely that token status through stereotype threat creates lower commitment, negative expectations toward future work situation and lower self-assessment. All expected to lead to lower management aspirations not just in own organizations but overall. As shown in Figure 1, however, most employees do not want to become a manager,³ and the distribution of answers is very right-skewed. In all, 43.6% of the respondents answered “not at all.” This indicates that the most interesting information in the variable is the difference between those who are not at all interested and those who are somewhat interested. Therefore, the variable was transformed into a dummy variable: 1 recoded into 0 = not interested at all, and >1 into 1 = interested.

Figure 1.

Respondent distribution regarding degree of interest in management position (percentage).

Because of this variable transformation, we use logistic regression to test our hypotheses. However, to test the robustness of our findings based on the above recoding, we did test our statistic models on different codings of the dependent variable.⁴ The robustness tests showed no significant differences in our findings.

Explanatory variable of main interest: Token status

To measure our key explanatory variable, token status, we asked the respondent the following: “How—approximately—is the allocation of male and female colleagues at your workplace? (If you are employed at a very large workplace divided into separate departments, think about the department or entity of employees that you are a part of).” The respondent could answer the following:

Approximately as many women as men,

Somewhat more women than men (approximately 60%-75%),

Many more women than men (more than 75%),

Somewhat more men than women (approximately 60%-75%), and

Many more men than women (more than 75%).

Each respondent’s status as token or not was then coded by combining information on respondent’s gender and whether or not they had ticked off Option 3 (more than 75% women), respectively, option 5 (more than 75% men). Hence, token status is defined as more than 75% of one’s colleagues being of the other gender, a definition which is slightly different from Kanter’s definition of a token as someone who is part of a group, making up less than 15% of the total workplace population, but in line with other studies of tokenism and diversity (see, for example, meta-study of Budig, 2002; Webber & Donahue, 2001). By using a cutoff point at 25%, our study constitutes a conservative test of the hypotheses, as one might argue that a cutoff point at 25% makes it even more difficult for potential effects of token status to manifest themselves in the statistical analysis. If we find token effects at a cutoff point of 25%, we can therefore be quite confident to also find effects at a 15% cutoff point. The down-side of this operationalization is that it, on the contrary, makes it a lenient test of hypothesis of nonsignificant correlations.

Furthermore, among theoretical and empirical scholars of tokenism, the entity within which token status is defined varies. Budig (2002) studies tokenism among occupations and industries with data aggregated at national level (Budig, 2002), while Yoder and Sinnett (1985) study everyday work-entity disregarding occupation and education, and Ott (1989) defines token status as professional peers within same work-entity. As mentioned above, we defined it as “colleagues at your work-place,” which makes the understanding of both “colleagues” and “work-place” subjective to the respondent. We argue that as the effect of token status to management aspirations is a social-psychological mechanism, it seems reasonable to use a measure that takes into account that people might regard who their colleagues and work place are differently.

Controls

Token status is of cause not the only factor determining career aspirations. To the contrary, theory and empirical research from several disciplines indicate the complex nature of career aspirations. Economists, often referring back to the work of Becker (1985), typically account for differences between the work orientations and career aspirations of men and women on the basis of job seekers’ utility maximizing choices, and they find that women with responsibilities in the family often choose to allocate less effort and commitment to their jobs than do men with similar skill levels and labor market experience. Structurally oriented sociologists typically begin with the assumption that “workers’ location in social structures affects their work attitudes and behavior because location signals whether career advancement is possible, and workers react accordingly” (Reskin & Bielby, 2005, p. 79). In accordance with this assumption, Cassirer and Reskin (2000) find that gender is not associated with promotion aspirations when the analyses control for opportunity structures. In other words, Cassirer and Reskin find that the differences between men’s and women’s career aspirations are owing to different structural opportunities for women and men when they are placed in a gender-segregated labor market. However, looking at Canadian students’ career aspirations, Schweitzer et al. (2011) find that gender is very significant.

The associations we hypothesize between token status and career aspirations are relatively independent of overall perspectives on career aspirations and career choices, and our explorations are not aimed at invalidating any of these. Our ambition is to explore the possible importance of an independent variable (gender token status in the workplace) for men’s and women’s career aspirations and not to explain the total variation in management aspirations (a dependent variable study). In our analyses, however, we do control for family situation and structural opportunities (through occupation-dummies) to construct the analyses adequately and subsequently to discuss the different overall perspectives on career aspirations, as we must control for variables that might covary with gender token status and management aspiration, either leading to over- or underestimation of the effect of token status if not included.

Below we discuss the rationale for the included control variables.

As the number of management positions differs within the different occupations included in the survey data and token status also differs within the occupations, we opt for dummy variables to control for each of the different occupations (descriptions of management aspirations by occupation are presented in Table A1 in the appendix).

Furthermore, we argue that some personal life situation characteristics affect both one’s management aspirations and tendency to choose a gender-untraditional workplace, and hence status as a token. Such personal life situation characteristics are thought to be age, children living at home, living with a partner, and length of employment. The argument is that the more well-educated and the less established your life situation is (e.g., not living with a partner or not having children), the more likely you are both to have management aspirations and to work in a job with token status. There might be gender differences here, however, such that these controls are more important regarding the correlation between token status and management aspirations for women than for men. Finally, we include working hours, measured as working (almost) full-time: 32 hr/week or more versus part-time work, because one might argue that choosing part-time work is an explicit—and to our analysis earlier—rejection of the aspiration to become a manager and that there is traditionally a high correlation between at least some highly gender-segregated occupations and part-time work (e.g., home carers, cleaners, and nurses).

All of the information on the controls is gathered from the items in the conducted survey. Descriptive statistics of all variables are presented in Table 2.

Table 2.

Descriptive Statistics.

Variable	Metric	All	Men	Women	Significant difference
Variable	Metric	M (SD)	M (SD)	M (SD)	Significant difference
Dependent
Managerial aspirations	0 = not interested; 1 = interested	0.44 (0.50)	0.49 (0.50)	0.41 (0.49)	***
Independent
Token status	0 = less than 75% of colleagues are of the other sex; 1 = 75% or more of colleagues are of the other sex	0.19 (0.39)	0.32 (0.47)	0.10 (0.30)	***
Controls
Age		43.42 (11.79)	44.12 (12.16)	42.78 (11.41)	***
Occupation	Dummy variables measuring each of the 13 different categories of occupations. Professional caretakers in 24-hr care institutions for vulnerable children and youths are reference category in the statistics.	See Table 1 for information
Working hours	Dummy variable 0 = part-time (= or ⩽ 32 hr a week); 1 = (almost) full-time (>32 hr a week)	0.81 (0.39)	0.88 (0.33)	0.76 (0.43)	***
Children living at home	0 = no; 1 = yes	0.51 (0.50)	0.47 (0.50)	0.54 (0.50)	***
Living with partner	0 = no; 1 = yes	0.76 (0.43)	0.74 (0.44)	0.78 (0.42)	**
Length of employment	Years of employment	9.34 (10.14)	9.40 (10.38)	9.32 (9.92)

p = .1. **p < .05. ***p < .01 (two-tailed).

Statistical Modeling

To test the hypotheses, we conduct the logistic regression analyses as separate models by gender to see if token status affects men and women differently.⁵ We conduct each of the statistics in three steps. In the first step—shown in Table 3 as Model 1s—we test the noncontrolled correlation between token status and management aspirations. In the next step—Model 2s—we add the variable measuring characteristics of personal life situation into the equation simply to see if personal life situation differences cover up, enhance, or minimize the apparent correlation. In the third step, we include the occupation dummies. By doing so, we explore whether or not the initial differences between occupations in relation to management aspirations and token status cover up the actual (cross-occupation) correlation between token status and management aspirations—both within the group of male and female employees and between them. Because male and female employees are tokens in different occupations, the association between token status and management aspirations might change differently when we control for occupations. Hence, the test of the hypotheses and exploration of potential nuanced differences between the correlation between token status and management aspirations lay in the significance of token status and how it changes between the three models when comparing males and females.

Table 3.

The Effect of Gender and Token Status on Employees’ Interest in Management Position.

	Interest in management position: no or yes
	Women			Men
	Model 1	Model 2	Model 3	Model 1	Model 2	Model 3
Constant	−0.414*** (0.661)	2.606 (13.551)	1.902*** (6.701)	−0.025 (0.975)	1.829*** (6.227)	1.912*** (6.77)
Token status	0.448** (1.565)	−0.061 (0.941)	−0.535** (0.586)	−0.071 (0.931)	0.117 (1.124)	0.202 (1.224)
Age		−0.069*** (0.934)	−0.058*** (0.944)		−0.068*** (0.935)	−0.067*** (0.935)
Working hours		0.479*** (1.615)	0.368* (1.183)		0.791*** (2.206)	0.754*** (2.127)
Children living at home		0.079 (1.082)	0.065 (1.067)		0.310* (1.364)	0.353** (1.423)
Living with partner		−0.264 (0.768)	−0.291* (0.748)		0.417** (1.517)	0.417** (1.517)
Length of employment		−0.024*** (0.976)	−0.026*** (0.974)		−0.001*** (0.999)	−0.000* (1.000)
Social and health workers			0.032 (1.033)			−0.085 (0.919)
Teachers in primary schools			0.335 (1.399)			−0.495 (0.609)
Physicians			1.511*** (4.532)			0.414 (1.513)
Health care professionals			0.089 (1.093)			−0.442 (0.642)
Office and IT staff			0.371 (1.450)			0.227 (1.255)
Academic staff in public administration			1.673*** (5.332)			−0.143 (0.867)
Professional caretakers in daycare institutions			−0.241 (0.786)			−0.440 (0.644)
Technical staff and cleaning			−0.028 (0.972)			0.157 (1.170)
Teachers in youth educations			0.618* (1.857)			−0.207 (0.813)
Researchers			1.358*** (3.889)			0.876* (2.402)
Police and prison staff			0.718* (2.049)			−0.853 (0.426)
Employees in the armed forces			1.536*** (4.647)			−0.344 (0.709)
Model statistic
N	1,181	1,181	1,181	820	820	869
Nagelkerke R²	.01**	.24***	.31***	.00	.23***	.26***

Note. Interest in Management position: no-yes logistic regression, cell entries are B-coefficients with odds ratios in parentheses.

p = .1. **p < .05. ***p < .01 (two-tailed).

The findings of the statistics are shown and discussed in the next section.

Findings

Analyses of data show that, on average, management aspirations are more widespread among men than among women. Almost half of the male respondents (49%) express some degree of interest in a management position compared with only 41% of the women. In addition, in the public sector—at least in Denmark—more men than women are tokens. In our data, only 10% of the women are tokens, whereas this figure is 33% for the men. This is no surprise, as most public employees in Denmark are found in traditional female sectors such as the health, caregiving, and teaching and as the public sector generally attracts more women than men.

The question remains, however, whether or not status as token is negatively correlated with management aspirations among all employees regardless of gender and occupation (H1); negatively correlated with management aspirations among women, but positively among men (H2); or is not correlated with career aspirations, but career aspirations will be more widespread among men than among women (H3). The results of this analysis are shown in Table 3.

The first and most important finding to note is that, controlled for alternative explanations with respect to both personal life situation and differences between occupations, token status is significantly negatively correlated with management aspirations among women, while the correlation is insignificant among men. In other words, we find only partial support for H2, namely the glass-ceiling effect for women, but no support for the glass-escalator effect for men.

However, the changes across the three model specifications and comparison of the changes between the statistics for men and women raise several intriguing questions. First, why does the correlation between token status and management aspirations among women change from significantly positive in Model 1 to insignificantly negative when we control for personal life situation (Model 2), whereas it remains insignificant for men in all models? Second, why does the insignificantly negative correlation between token status and management aspirations among women become fairly large and significant when we control for occupations?

Controlling for personal life situation influences the correlation between token status and management aspirations differently among women, because token men and token women have different profiles when it comes to personal life situation compared with nontokens of same gender. A closer look at the differences in mean between token status and each of the personal characteristics within the group of men and women, respectively, reveals interesting differences (see Table 4).

Table 4.

Differences in Means Between Token Versus Nontoken Men, and Token Versus Nontoken Women.

Variable	Women			Men
	Token	Nontoken	Significant difference	Token	Nontoken	Significant difference
	M (SD)	M (SD)	Significant difference	M (SD)	M (SD)	Significant difference
Managerial aspirations	0.51 (0.50)	0.40 (0.49)	**	0.48 (0.50)	0.49 (0.50)
Age	36.61 (11.51)	43.43 (11.22)	***	45.03 (12.33)	44.12 (12.04)
Occupation	—	—	—	—	—	—
Working hours	0.95 (0.21)	0.74 (0.44)	***	0.86 (0.44)	0.89 (0.31)
Children living at home	0.48 (0.50)	0.55 (0.50)	*	0.44 (0.50)	0.49 (0.50)
Living with partner	0.66 (0.48)	0.79 (0.41)	***	0.74 (0.44)	0.75 (0.43)
Length of employment	10.04 (9.90)	9.23 (9.92)	*	8.54 (10.07)	8.91 (10.08)	***

p = .1. **p < .05. ***p < .01 (two-tailed).

First, token males are only different from nontoken males with respect to length of employment. Token men have been employed for a shorter time than nontoken men. This finding suggests two possible explanations: First, token men might work in a sector for a short while—for example, the elder or child care sector. They work there for a year or so—perhaps referred by the employment service—before they find a “real man’s job.” Second, token men might generally have—to a higher degree than token women—a tendency to leave jobs and organizations where they are tokens, simply because they—to a lesser degree than women—submit to or enjoy the token status. In contrast, women with token status are different from nontoken women on several personal characteristics. They are younger, less likely to live with a partner, less likely to have children living at home, and have been employed for longer than female nontokens. Hence, token women fit two predominant caricatures of token women, namely the young, single career woman or the mature, childless woman; two caricatures that are not mutually exclusive, as the young career woman might one day become a mature, childless woman. Put together, the personal life profiles of token women and token men vary differently from those of nontokens of same gender. However, token women vary more from nontoken women than token men do from nontoken men. This might be why the correlation between token status and management aspirations changes drastically among women controlling for personal life situation, while it only changes insignificantly among men.

Furthermore, examining Table 3, we also notice that within the group of women, there are four occupations (physicians, academic staff in public administration, researchers, and employees in the armed forces) where the membership of the occupation in itself not only is positively correlated with management aspirations, but the correlation coefficients exceed a possible negative correlation with being token. Within the male group, this only occurs in the case of researchers and with a reversed correlation for men in the health care occupation. This shows that almost regardless of occupation, men have high levels of management aspirations (relative to women), while the diversity based on occupation is much higher among women.

However, this does not explain why the insignificantly negative correlation between token status and management aspirations among women (in Model 2) increases and becomes significant when we control for occupations. One explanation could be that one or some of the occupations act as a suppressing variable due to a high correlation with both token status and management aspiration. In other words, the occupations with a high level of token status are also those with the largest share of female employees interested in management positions—hence, the isolated effect of token status only appears when we control for these occupations. To untangle this, we identified the occupations where female token status is found significantly more frequently than within the rest of the occupations and where the female tokens have a significantly high level of management aspiration: research and the armed forces.⁶ After having identified these occupations, we ran a fourth regression (not shown) without these two occupations. Comparison of this new (fourth) regression and the one in Model 3 in Table 3 uncovers that the negative correlation between token status and management aspirations only becomes significant when adding the two occupations. In other words, women in research and the armed forces are very often token women—more often than women in other public occupations—but at the same time they are more likely to express management aspirations than women in other occupations. To which degree this is an occupation selection or a value-learning effect is difficult to determine; it might very well be a mix of both. The results are, however, very much in line with Kanter’s analysis of Indsco, in which she found a correlation between female tokenism and preferences for power and career opportunities (Kanter, 1977a). Without controlling for these occupations, however, we suppress the negative correlation between token status and women’s management aspirations.

Thus, the conclusion is that token status has a negative correlation with management aspirations among women, but token women are at the same time mainly to be found in occupations where women have a relatively high level of managerial aspirations—and to such a degree that their token status does not completely eliminate their interest in management. One might therefore say that token women are attracted to management, but their interest might be even more pronounced if they were not in a gender minority position in their workplace.

Discussion

In this article, we have explored associations between gender, gender token status, and management aspirations among employees in 13 public sector occupations in Denmark. We began by asking whether workplace token status affect management aspirations of women and men, whether it affects men and women alike, and whether the potential gendered effect of workplace token status are generalizable across occupational contexts.

We find that gender token status is negatively associated with management aspirations among women, while it is not associated with management aspirations among men. Furthermore, we find that, despite variation in women’s management aspirations across occupations, token status continues to restrain women’s aspirations, likewise is men’s management aspirations constantly high (relative to women’s) and unaffected by token status across different occupations. Hence, the gender-different effect of token status seems to be generalizable across occupational contexts.

In the following, we discuss our findings in detail.

As shown, our findings do not support our first hypothesis (H1), which, based on Kanter’s seminal token study from 1977, assumes that all token persons are negatively affected by their token status. However, the results partly supports our second hypothesis (H2) and indicates that token status in the workplace affects women and men differently. We suggest that the differences between male and female tokens can be explained by stereotype threat theory, according to which female tokens may be seen and judged according to a negative stereotype about women, work, and career, which negatively affects their self-assessments and dampen their career aspirations. Male tokens do not suffer from the same clash between gender stereotype and the demands typically associated with the manager role, and in our study, male tokens’ management aspirations are not affected in comparison with nontokens.

Our third hypothesis (H3) states that while management aspirations might be characterized by marked gender differences, these differences may not be associated with token status. Our findings lend some support to this hypothesis.

In our data, career aspirations are noticeably more pronounced among men than among women, irrespective of occupation, token status, and other controls. This finding supports Budig’s notion of “male advantage as usual,” Acker and Williams’s notion of the gendered organization, and Mainiero and Sullivans’s conception of gendered alpha and beta career patterns. The result indicates that men in general (not token men in particular) enjoy a structural advantage, helping them to career aspirations more readily than women.

However, among the women in the study, we find—as mentioned—that career aspirations vary widely across the different occupations. In four of the occupations in the study—physicians, academic staff in public administration, researchers, and employees in the armed forces—the positive effect of occupation on women’s career aspirations outweighs the negative effect of token status. In these four “female career occupations,” token status continues to restrain women’s career aspirations, but no more so than that the net result is heightened career aspirations compared with the average among female public employees in the study. Compared with typically female-dominated occupations, these four occupations represent better structural opportunities with respect to management career, well-defined career paths, career ladders, and so on. From a structural point of view, it is therefore not surprising to find more pronounced career aspirations among female employees in these occupations than among women in areas such as social and health workers, cleaning, or elementary school teaching, where career opportunities are more limited.

Because of the cross-sectional design of the study, obviously we cannot determine if women with relatively pronounced career aspirations self-select themselves into the four “female career occupations” or if the structural possibilities and other features of the social environment in the occupations work to promote or produce career aspirations among female employees. However, finding the negative effect of tokenism in these occupations indicates that the organizational environment makes a difference for the career aspirations among female employees. Ambitious women are very likely to self-select themselves into occupations with above-average career opportunities, but it does not seem likely that the negative effect of tokenism is a result of self-selection in the same manner. The effect of tokenism more likely indicates the importance of interactional patterns like stereotyping and stereotype threat in the workplace for employees’ career aspirations. To untangle the relative importance of self-selection and intra-organizational interactional patterns for employees’ career aspirations, however, further research based on panel design is required.

In contrast to the widely varying female career aspirations across the 13 occupations, there is—and mentioned above—a relatively constant and (in comparison with women) high level of managerial aspirations among men in the 13 very different occupations. Managerial aspirations among men thus seem to be relatively independent of the variation in occupation-specific career opportunities. When it comes to managerial aspirations, male employees thus seem to be much less sensitive than female employees to opportunity structures or occupation-specific gender cultures. This finding possibly indicates that career and authority aspirations are highly gender-typed male and that they are attached to maleness as a societal master status. It is also consistent with the notion of gendered alpha and beta career patterns. Career and management aspirations do not seem to be gender typified to the same extent for women. Based on the current study, one could argue that women’s career aspirations are in line with opportunity structures to a higher degree than those of men. The more constant career aspirations among men across different occupational settings (representing different objective structures of opportunity) can be interpreted in several ways. The relative consistency can be due to greater susceptibility to societal gender stereotypes among men than among women. However, it can also be due to persistent career advantages for men irrespective of occupational context.

The result that career aspirations are generally more pronounced among men than among women (net of occupation, working hours, tenure, and family situation) partly challenges Cassirer and Reskin’s (2000) conclusion that differences in aspirations for promotion between women and men disappear when you control for jobs with unequal opportunities.

In fact, our study both challenges and confirms Cassirer and Reskin’s findings. We find that opportunity structures (represented in our study by the 13 occupations) are significantly associated with women’s career aspirations; so much so that career aspirations are typically more pronounced among token women than among nontokens—because token women in the study are typically placed in occupational settings, displaying much better opportunity structures than the female-dominated occupations. As we have discussed, however, we find that career aspirations among male employees are relatively unaffected by token status as well as occupational setting. And this finding runs counter to the notion that opportunity structures tend to determine career aspirations.

As mentioned, the present study is not without limitations. Even though the Danish cross-sectional data allow us to partly control for confounding variables, partly rule out others based on characteristics of the Danish settings, we cannot jump from the observed correlations to a claim of causality, as there is a risk of omitted variable bias. Furthermore, despite comprehensive and forceful theoretical arguments, we cannot be certain that management aspirations are second to token status or vice-versa. Hence future studies might focus on putting the concluded correlations and thus deduced hypothetical explanations to a more thorough test by using instrumental variables or longitudinal data.

Furthermore, the definition of token status as a cutoff point of 25% minority makes it—as argued earlier—a conservative test of hypothesis expecting a significant correlation between token status and management aspirations. The down-side, however, is that this definition on the contrary makes it a lenient test of hypothesis about a nonsignificant correlation. Hence, further studies might focus on putting the findings regarding the in-significant correlations between men’s token status and management aspirations to a more conservative test.

Supplemental Material

korrelationsmatrix2 – Supplemental material for Token Status and Management Aspirations Among Male and Female Employees in Public Sector Workplaces

Supplemental material, korrelationsmatrix2 for Token Status and Management Aspirations Among Male and Female Employees in Public Sector Workplaces by Vibeke Lehmann Nielsen and Mikkel Bo Madsen in Public Personnel Management

Footnotes

Appendix

Table A1.

Descriptive of Management Aspirations by Occupations.

	Interest in management position: no-yes
Professional caretakers in 24-hr care institutions for vulnerable children and youths	0.37 (0.48)
Social and health workers	0.35 (0.48)
Teachers in primary schools	0.36 (0.48)
Physicians	0.75 (0.43)
Health care professionals	0.36 (0.48)
Office and IT staff	0.41 (0.49)
Academic staff in public administration	0.61 (0.49)
Professional caretakers in day care institutions	0.35 (0.48)
Technical staff and cleaning	0.35 (0.48)
Teachers in youth educations	0.38 (0.49)
Researchers	0.70 (0.46)
Police and prison staff	0.40 (0.49)
Employees in the armed forces	0.65 (0.48)

Note. Mean score and standard deviation in parentheses.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Vibeke Lehmann Nielsen, PhD, is a professor in the Department of Political Science, University of Aarhus. His research interests are street-level bureaucrats, gender and work behavior, client behavior, regulation and compliance.

Mikkel Bo Madsen, PhD, is an associate professor in the Department of Social Work, Metropolitan University College, Copenhagen, Denmark. His research interests are professions and professional work in the public sector, organization theory, gender and organization, motivation, employment policies, unemployment.

References

Acker

(1990). Hierarchies, jobs, bodies: A theory of gendered organizations. Gender & Society, 4, 139-158.

Atwater

L. E.

Van Fleet

D. D.

(1997). Another ceiling? Can males compete for traditionally female jobs? Journal of Management, 23, 603-626.

Becker

G. S.

(1985). Human capital, effort, and the sexual division of labor. Journal of Labor Economics, 3(1, Pt. 2), 33-58.

Bendl

Schmidt

(2010). From glass ceilings to firewalls: Different metaphors for describing discrimination. Gender, Work & Organization, 17, 612-634.

Berger

Ridgeway

C. L.

Zelditch

(2002). Construction of status and referential structures. Sociological Theory, 20, 157-179.

Block

C. J.

Koch

S. M.

Liberman

B. E.

Merriweather

T. J.

Roberson

(2011). Contending with stereotype threat at work: A model of long-term responses. The Counseling Psychologist, 39, 570-600.

Bloksgaard

(2011). Masculinities, femininities and work: The horizontal gender segregation in the Danish labour market. Nordic Journal of Working Life Studies, 2(1), 5-21.

Budig

M. J.

(2002). Male advantage and the gender composition of jobs: Who rides the glass escalator? Social Problems, 49, 258-277.

Cassirer

Reskin

B. F.

(2000). High hopes: Organizational position, employment experiences, and women’s and men’s promotion aspirations. Work and Occupations, 27, 438-463.

10.

Charles

(2003). Deciphering sex segregation: Vertical and horizontal inequalities in ten national labor markets. Acta Sociologica, 46, 267-287.

11.

Cognard-Black

A. J.

(2004). Will they stay or will they go? Sex atypical work among token men who teach. The Sociological Quarterly, 45, 113-139.

12.

Cohen

Swim

(1995). The differential impact of gender ratios on women and men. Personality and Social Psychology Bulletin, 21, 876-884.

13.

Correll

S. J.

(2001). Gender and the career choice process: The role of biased self-assessments. American Journal of Sociology, 106, 1691-1730.

14.

Correll

S. J.

(2004). Constraints into preferences: Gender, status, and emerging career aspirations. American Sociological Review, 69, 93-113.

15.

DiTomaso

Post

Parks-Yancy

(2007). Workforce diversity and inequality: Power, status, and numbers. Annual Review of Sociology, 33, 473-501.

16.

Driskell

J. E.

(1982). Personal characteristics and performance expectations. Social Psychology Quarterly, 45, 229-237.

17.

Driskell

J. E.

Olmstead

Salas

(1993). Task cues, dominance cues, and influence in task groups. Journal of Applied Psychology, 78, 51-60.

18.

Ely

R. J.

(1995). The power in demography: Women’s social construction of gender identity at work. Academy of Management Journal, 38, 589-634.

19.

Emerek

Holt

(2008). Lige muligheder—frie valg? Om det kønsopdelte arbejdsmarked gennem et årti [Equal opportunities - free choices? The gender segregated labor marked thoughout a decade]. Copenhagen, Denmark: SFI—The Danish National Centre for Social Research.

20.

Fairhurst

G. T.

Snavely

B. K.

(1983a). Majority and token minority group relationships: Power acquisition and communication. The Academy of Management Review, 8, 292-300.

21.

Fairhurst

G. T.

Snavely

B. K.

(1983b). A test of the social isolation of male tokens. Academy of Management Journal, 26, 353-361.

22.

Floge

L. L.

Merrill

D. M.

(1986). Tokenism reconsidered: Male nurses and female physicians in a hospital setting. Social Forces, 64, 925-947.

23.

Gans

J. E.

(1987). Men’s career advantages in nursing: The principle of the Peter. In Lopata

H. Z.

(Ed.), Current research on occupations and professions (Vol. 4, pp. 181-198). New York, NY: JAI Press.

24.

Gidengil

Vengroff

(1997). Representative bureaucracy, tokenism and the glass ceiling: The case of women in Quebec municipal administration. Canadian Public Administration, 40, 457-480.

25.

Greed

(2000). Women in the construction professions: Achieving critical mass. Gender, Work & Organization, 7, 181-196.

26.

Gustafson

J. L.

(2008). Tokenism in policing: An empirical test of Kanter’s hypothesis. Journal of Criminal Justice, 36, 1-10.

27.

Hammond

J. A.

Mahoney

C. W.

(1983). Reward-cost balancing among women coalminers. Sex Roles, 9, 17-29.

28.

Heikes

E. J.

(1991). When men are the minority: The case of men in nursing. The Sociological Quarterly, 32, 389-401.

29.

Heilman

M. E.

(2012). Gender stereotypes and workplace bias. Research in Organizational Behavior, 32, 113-135.

30.

Hultin

(2003). Some take the glass escalator, some hit the glass ceiling? Career consequences of occupational sex segregation. Work and Occupations, 30, 30-61.

31.

Kanter

R. M.

(1977a). Men and women of the corporation. New York, NY: Basic Books.

32.

Kanter

R. M.

(1977b). Some effects of proportions on group life: Skewed sex ratios and responses to token women. American Journal of Sociology, 82, 965-990.

33.

King

E. B.

Hebl

M. R.

George

J. M.

Matusik

S. F.

(2010). Understanding tokenism: Antecedents and consequences of a psychological climate of gender inequity. Journal of Management, 36, 482-510.

34.

Konrad

A. M.

Kramer

Erkut

(2008). Critical mass: The impact of three or more women on corporate boards. Organizational Dynamics, 37, 145-164.

35.

Madsen

M. B.

Holt

Jonassen

A. B.

Schademan

H. K.

(2010). Kvinder og mænd i den offentlige sektor. Karrieremønstre, lederønsker og ledermuligheder [Women and men in the public sector. Career patterns, management aspirations and management opportunities]. Copenhagen, Denmark: SFI—The Danish National Centre for Social Research.

36.

Mainiero

L. A.

Sullivan

S. E.

(2006). The opt-out revolt: Why people are leaving companies to create kaleidoscope careers. Mountain View, CA: Davies-Black.

37.

Mariani

M. D.

(2008). A gendered pipeline? The advancement of state legislators to congress in five states. Politics & Gender, 4, 285-308.

38.

Maume

D. J.

(1999). Glass ceilings and glass escalators: Occupational segregation and race and sex differences in managerial promotions. Work and Occupations, 26, 483-509.

39.

McDonald

T. W.

Toussaint

L. L.

Schweiger

J. A.

(2004). The influence of social status on token women leaders’ expectations about leading male-dominated groups. Sex Roles, 50, 401-409.

40.

McIlwee

J. S.

Robinson

J. G.

(1992). Women in engineering: Gender, power, and workplace culture. Albany: State University of New York Press.

41.

E. S.

Wiesner

W. H.

(2007). Are men always picked over women? The effects of employment equity directives on selection decisions. Journal of Business Ethics, 76, 177-187.

42.

Ott

E. M.

(1989). Effects of the male-female ratio at work: Policewomen and male nurses. Psychology of Women Quarterly, 13, 41-57.

43.

Rashotte

L. S.

Webster

Jr. (2005). Gender status beliefs. Social Science Research, 34, 618-633.

44.

Reskin

B. F.

(1993). Sex segregation in the workplace. Annual Review of Sociology, 19, 241-270.

45.

Reskin

B. F.

Bielby

D. D.

(2005). A sociological perspective on gender and career outcomes. Journal of Economic Perspectives, 19, 71-86.

46.

Reskin

B. F.

McBrier

D. B.

Kmec

J. A.

(1999). The determinants and consequences of workplace sex and race composition. Annual Review of Sociology, 25, 335-361.

47.

Ridgeway

C. L.

(1991). The social construction of status value: Gender and other nominal characteristics. Social Forces, 70, 367-386.

48.

Ridgeway

C. L.

(2001). Inequality, status, and the construction of status beliefs. In Turner

J. H.

(Ed.), Handbook of sociological theory (pp. 323-340). New York, NY: Kluwer Academic.

49.

Ridgeway

C. L.

Johnson

Diekema

(1994). External status, legitimacy, and compliance in male and female groups. Social Forces, 72, 1051-1077.

50.

Roberson

Deitch

E. A.

Brief

A. P.

Block

C. J.

(2003). Stereotype threat and feedback seeking in the workplace. Journal of Vocational Behavior, 62, 176-188.

51.

Roberson

Kulik

C. T.

(2007). Stereotype threat at work. Academy of Management Perspectives, 21(2), 24-40.

52.

Roth

L. M.

(2004). The social psychology of tokenism: Status and homophily processes on Wall Street. Sociological Perspectives, 47, 189-214.

53.

Sachdev

Bourhis

R. Y.

(1987). Status differentials and intergroup behaviour. European Journal of Social Psychology, 17, 277-293.

54.

Schweitzer

Lyons

Kuron

(2011). Exploring the career pipeline: Gender differences in pre-career expectations. Relations Industrielles/Industrial Relations, 66, 422-444.

55.

Spangler

Gordon

M. A.

Pipkin

R. M.

(1978). Token women: An empirical test of Kanter’s hypothesis. American Journal of Sociology, 84, 160-170.

56.

Spencer

S. J.

Steele

C. M.

(1999). Stereotype threat and women’s math performance. Journal of Experimental Social Psychology, 35(1), 4-28.

57.

Steele

C. M.

Aronson

(1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69, 797-811.

58.

Steele

C. M.

Spencer

S. J.

Aronson

(2002). Contending with group image: The psychology of stereotype and social identity threat. Advances in Experimental Social Psychology, 34, 379-440.

59.

Tomaskovic-Devey

Zimmer

Stainback

Robinson

Taylor

McTague

(2006). Documenting desegregation: Segregation in American workplaces by race, ethnicity, and sex, 1966-2003. American Sociological Review, 71, 565-588.

60.

Turco

C. J.

(2010). Cultural foundations of tokenism: Evidence from the leveraged buyout industry. American Sociological Review, 75, 894-913.

61.

Webber

S. S.

Donahue

L. M.

(2001). Impact of highly and less job-related diversity on work group cohesion and performance: A meta-analysis. Journal of Management, 27, 141-162.

62.

Williams

C. L.

(1992). The glass escalator: Hidden advantages for men in the “female” professions. Social Problems, 39, 253-267.

63.

Williams

C. L.

(1995). Still a man’s work: Men who do “women’s work.” Berkeley: University of California Press.

64.

Yap

Konrad

A. M.

(2009). Gender and racial differentials in promotions: Is there a sticky floor, a mid-level bottleneck, or a glass ceiling? Relations Industrielles/Industrial Relations, 64, 593-619.

65.

Yoder

J. D.

(1985). An academic woman as a token: A case study. Journal of Social Issues, 41(4), 61-72.

66.

Yoder

J. D.

(1991). Rethinking tokenism: Looking beyond numbers. Gender & Society, 5, 178-192.

67.

Yoder

J. D.

(1994). Looking beyond numbers: The effects of gender status, job prestige, and occupational gender-typing on tokenism processes. Social Psychology Quarterly, 57, 150-159.

68.

Yoder

J. D.

(2002). 2001 Division 35 presidential address: Context matters: Understanding tokenism processes and their impact on women’s work. Psychology of Women Quarterly, 26, 1-8.

69.

Yoder

J. D.

McDonald

T. W.

(1998). Measuring sexist discrimination in the workplace: Support for the validity of sexist events. Psychology of Women Quarterly, 22, 487-491.

70.

Yoder

J. D.

Sinnett

L. M.

(1985). Is it all in the numbers? Psychology of Women Quarterly, 9, 413-418.

71.

Young

J. L.

James

E. H.

(2001). Token majority: The work attitudes of male flight attendants. Sex Roles, 45, 299-319.

72.

Zimmer

(1988). Tokenism and women in the workplace: The limits of gender-neutral theory. Social Problems, 35, 64-77.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB