Managerial stereotypes over time: the rise of feminine leadership

Abstract

Purpose

A key obstacle to women’s advancement to managerial roles and leadership positions is the stereotype of the “good” manager, which is characterized by masculine traits. Although this gendered managerial stereotype has been very persistent over the past decades, Powell et al. (2021) recently showed that business students in the USA reported a decreased preference for masculine leadership traits and an increased preference for feminine leadership traits, resulting in a so-called “androgynous” manager profile that contains both masculine and feminine characteristics. This study aims to replicate Powell et al.’s (2021) findings among an older sample of working adults in The Netherlands.

Design/methodology/approach

The present study tests for changes in the managerial stereotype in a sample of 5,542 Dutch employees across 2005, 2010 and 2020.

Findings

In line with Powell et al. (2021), the results confirm employees’ decreased preference for masculine and increased preference for feminine leadership traits in 2020 compared to 2005. Nevertheless, Dutch employees still favored masculine over feminine leadership traits in 2020, contrary to the findings by Powell et al. (2021).

Practical implications

These observed changes in the managerial stereotype could prove to be an important step forward for women’s advancement to management and leadership positions.

Originality/value

With the present study, the authors demonstrate cross-cultural generalizability and conclude that the stereotype of a “good manager” is not only changing among US business students but also among working adults in The Netherlands. Overall, this study strengthens the observation that the stereotype of a “good manager” is becoming less gendered.

Citation

Feenstra, S., Stoker, J.I., Lammers, J. and Garretsen, H. (2023), "Managerial stereotypes over time: the rise of feminine leadership", Gender in Management, Vol. 38 No. 6, pp. 770-783. https://doi.org/10.1108/GM-10-2022-0331

Despite growing numbers of women advancing to management positions, globally such positions are still dominated by men [Organization for Economic Co-operation and Development (OECD), 2022]. Extensive research has examined the reasons for the lack of advancement of women to management positions (Ellemers, 2014; Hoyt, 2010; Lyness and Thompson, 2000). A key obstacle that this line of research has identified is the stereotype of the “good” manager, which is characterized by masculine traits (e.g. be assertive, dominant and powerful) rather than feminine traits (e.g. be compassionate, warm and friendly; for a review see Koenig et al., 2011). Because of this stereotype, women are often portrayed as lacking managerial qualities, while men are seen as having a more natural fit with management positions (Eagly and Karau, 2002). This, in turn, hinders women’s advancement to powerful positions (Heilman, 2001).

For a long time, the gendered managerial stereotype was remarkably stable. For instance, Powell et al. (2002), asked respondents over several decades to rate the degree to which they thought masculine and feminine traits were characteristic of a good manager. Their results showed that in 1979, 1989 and 2002, people preferred masculine over feminine leadership traits. Moreover, few to no changes in preferences for these traits were observed over time. Stoker et al. (2012) found comparable results among Dutch employees in 2005, with these employees preferring masculine over feminine leadership traits.

A meta-analysis by Koenig et al. (2011) on managerial stereotypes painted a more optimistic picture. Although the results of their meta-analysis confirmed the gendered managerial stereotype, with people favoring masculine over feminine traits in managers, their findings also suggested that these stereotypes are changing. In fact, the results of their meta-analysis indicated that the more recent studies in the meta-analysis showed a decreased preference for masculine over feminine leadership traits. This observation is supported by recent empirical research. Specifically, Powell et al. (2021) observed that business students favored feminine leadership traits more and masculine leadership traits less in 2018 compared to the previous decades, resulting in an “androgynous” profile, for which no preference for masculine over feminine leadership traits was observed.

Although the recent work by Powell et al. (2021) is valuable and paints an optimistic picture of changing managerial stereotypes, two issues remain to be addressed to increase our confidence in the reported developments. First, Powell et al. (2021) examined the leadership perceptions of US business students. Although 30% of Powell et al.’s (2021) sample were part-time MBA students and therefore had substantial work experience, it is important to generalize these findings beyond business students. In fact, the meta-analysis by Koenig et al. (2011) showed that the majority of research examining managerial stereotypes has been conducted using student samples. Moreover, the findings of their meta-analysis suggest that students hold slightly different stereotypes of leaders compared to older participants with more work experience. Considering these findings, we believe it is important to examine whether working adults show a similar decline in preference for masculine over feminine leadership traits as the US business students in Powell et al.’s (2021) sample.

Second, Powell et al. (2021) only examined managerial stereotypes among students located in the USA. Therefore, it is unclear whether these findings generalize to other countries and cultures. In this regard, the USA has seen a rapid increase of women entering the workforce and gaining positions of power over the past decades [International Labour Office (ILO), 2020; Scarborough, 2018]. As such, the USA has transitioned from a relatively masculine culture with dominant gender roles (Hofstede, 1980) to a feminine culture where feminine qualities are deemed more important (Wu, 2006). The Netherlands, where the preference for masculine characteristics was also found, has traditionally been a feminine culture and has thus undergone less of a change in this regard (Hofstede, 1980). It therefore remains an open question whether or not the reported changes in managerial stereotypes in a US sample generalize to traditional feminine cultures such as The Netherlands.

In fact, Powell et al. (2021, p. 726) stated that their “results may be unique to these two populations [US business students] and should not be generalized to other populations or national cultures without additional research. Testing of their external validity is recommended.” Many scholars have argued that such replication is a fundamental part of scientific progress and strengthens our confidence in the observed trends (Asendorpf et al., 2016; Eden, 2002; Jasny et al., 2011). The purpose of this study, therefore, is to replicate the findings of Powell et al. (2021) in a sample of working adults in Europe across a large time-span. We compare 5,542 Dutch employees’ preferences for masculine and feminine leadership traits in 2005 (reported in Stoker et al., 2012), 2010 and 2020 and examine whether the stereotype of what constitutes a good manager has changed over time. In doing so, we aim to replicate previous findings of changes in managerial stereotypes among US business students among an older sample of Dutch employees.

Theory

Managerial stereotypes

Stereotypes are defined as “beliefs about the characteristics, attributes, and behaviors of members of certain groups” (Hilton and Von Hippel, 1996, p. 240). Stereotypes describe not only what members of certain groups or social categories are like (i.e. descriptive stereotypes) but also what people feel that members of certain groups should be like (i.e. prescriptive stereotypes; Burgess and Borgida, 1999). These, descriptive and prescriptive stereotypes often overlap, meaning that people who deviate from what is descriptively expected based on their group membership evoke negative feelings due to a violation of expectations (Mendes et al., 2002; Rudman and Glick, 2001; Rudman and Phelan, 2008). This also means that people may form implicit expectations about the extent to which different categories (such as gender and leadership) are compatible or mutually exclusive.

In the early 1970s, Schein (1973, 1975) identified a masculine stereotype for the role of manager. In her work, she asked male and female middle managers to rate women in general, men in general, and successful middle managers using a list of 92 adjectives. The results showed a strong similarity between the ratings of successful middle managers and the ratings of men in general, but not between the ratings of successful middle managers and ratings of women in general. In other words, successful managers were rated as more similar to men than to women. This effect was later replicated in various countries and among various samples, and the method to show it is often referred to as the think manager – think male paradigm (for an overview, see Schein, 2001).

One theoretical framework that can explain this effect is role congruity theory (Eagly and Karau, 2002). Combining social role theory (Eagly, 1987; Eagly et al., 2000) with models of the structure of social perception (Abele and Wojciszke, 2007; Fiske et al., 2018), role congruity theory describes that gender stereotypes include communal and agentic attributes. More specifically, the theory proposes that people expect women to have communal characteristics (e.g. be helpful, kind and sympathetic) and men to have agentic characteristics (e.g. be assertive, controlling and confident). These stereotypes of what men and women are and should be relate directly to the stereotype that people have of a “good manager”, with people perceiving such a “good manager” to be agentic, but not communal. As a result, the stereotype for men overlaps with the stereotype for leaders, while the stereotype for women does not. As a result, men are seen as having a “natural fit” with leadership and holding managerial roles, while women are not (or less so).

Powell et al. (2002); Powell and Butterfield (1979, 1989, 2015) empirically examined this gendered managerial stereotype. In their research, they asked students to rate a good manager on communal and agentic attributes, using the Bem Sex Role Inventory (BSRI; Bem, 1974). Bem used the words “masculine” and “feminine” instead of “agentic” and “communal”, respectively. In line with the masculine managerial stereotype, results showed that students consistently favored masculine over feminine leadership traits. The same results were found for employees in 2005 by Stoker et al. (2012). Assuming that these traits are representative for the concepts of male and female leadership, this line of research thus shows a strong preference for masculine leadership characteristics, meaning that individuals prefer masculine over feminine leadership traits in managers. Building on this line of work, we propose the following:

H1.

Employees value masculine over feminine leadership characteristics.

Changes in managerial stereotypes

Although stereotypes are “enduring human phenomena” (Fiske, 1998, p. 357), the content of stereotypes is subject to change. In particular, stereotypes of women and men have changed substantially over the past decades (Charlesworth and Banaji, 2022; Eagly et al., 2020). This change is consistent with the central argument in social role theory which states not only that men and women behave according to gender stereotypes, but also that these gender stereotypes are inferred from the way men and women typically behave (Eagly, 1987; Eagly et al., 2000). This implies that if the typical behavior of men and women in society changes, gender stereotypes may follow. Reflecting the increase of women in higher-status positions, the belief that women are equally competent as men has increased substantially over time. In contrast, stereotypical differences in agency have changed much less (Eagly et al., 2020).

Similar to gender stereotypes, the ideas and expectations about what people consider a “good manager” may have been subject to change as well. Researchers suggest that such trends reflect changes in society in general and in the workplace specifically (Offermann and Coats, 2018). Over the past decades, there have been two notable developments in the workplace that likely influenced the managerial stereotype.

First, an increasing number of women have broken the glass ceiling and entered leadership and managerial roles. A recent report (Catalyst, 2019), for example, observed a considerable growth of women in managerial positions across the globe in the period from 1991 to 2018. Moreover, the report showed that this increase was strongest in Asia and the Pacific, Latin America and the Caribbean and Europe and Central Asia and became steeper from 2002 onwards. More recent reports from the Organization for Economic Co-operation and Development [Organization for Economic Co-operation and Development (OECD), 2022] and the International Labour Organization [International Labour Office (ILO), 2020] showed similar trends.

Second, over the past two decades, the workplace has undergone major changes with respect to its focus on, and appreciation of, traditionally feminine and masculine qualities and behaviors (Blake-Beard et al., 2020; Gerzema and D’Antonio, 2013). Due to globalization and technological developments, organizations have become more complex, focusing more on teamwork and shifting from predominantly autocratic, masculine, leadership styles, to more democratic, feminine leadership styles (Eagly and Carli, 2003; Gergen, 2005). This shift in focus is also reflected in leadership theories and frameworks that increasingly center on feminine (as opposed to masculine) traits and qualities (e.g. transformational leadership, servant leadership and ethical leadership; Bass and Rigio, 2006; Brown et al., 2005; Russell, 2001).

Although numerous researchers have argued for changes in the managerial stereotype, empirical evidence for such a change is scarce (Offermann and Coats, 2018; Powell et al., 2002). In a study by Stoker et al. (2012), however, it was confirmed that the stereotype of a “good manager” was different for respondents who had a female manager or who worked in an organization with a relatively high percentage of female leaders. These respondents showed comparable preferences for masculine and feminine traits. Similarly, a meta-analysis on the masculine leader stereotype by Koenig et al. (2011) suggested that the masculine construal of leadership has decreased over time. They reviewed empirical papers (ranging from 1979 to 2007) examining the managerial stereotype and found that the difference between agentic and communal scores was smaller in articles published in later years.

In an effort to empirically confirm this observation, Powell et al. (2021) compared the rating of agentic and communal leadership traits collected in 2018 with similar data collected during the previous four decades. After almost 40 years of stability, they found a change in the stereotype in 2018: business students favored feminine leadership traits more and masculine leadership traits less, ultimately resulting in a stereotype where students equally include feminine and masculine leadership traits. To build on these findings from Powell et al. (2021), we hypothesize the following for respondents in an organizational setting:

H2.

Employees value masculine over feminine leadership characteristics less in the past decade.

Method

Procedure and participants

We reached out to readers of a Dutch management platform/magazine in 2005 (also reported in Stoker et al., 2012), 2010 and 2020, with a request to participate in a survey on leadership. As the survey asked about workplace experiences, only individuals who were employed at the time of study were eligible to participate. In total, 5,541 participants completed our measure of feminine and masculine leadership (Mage = 38.20, SD = 8.46; 41% female) [1]. See Table 1 for detailed characteristics of the three samples.

Measures

Unless indicated otherwise, participants responded to all items below on a scale from 1 (not at all) to 5 (very much).

Time.

We created two dummy-coded time variables and used 2005 as the reference category. Dummy 2010 was coded 1 for 2010 and 0 for 2005 and 2020. Dummy 2020 was coded 1 for 2020, and 0 for 2005 and 2010.

Gender.

Gender was coded 0 for female participants, 1 for male participants and 2 for other.

Ideal leadership characteristics.

We measured the stereotype of a “good manager” with a short version of the Bem Sex Role Inventory (BSRI, Bem, 1974). We showed participants a list of leadership qualities and asked them to indicate the degree to which they felt traits were characteristic of a good manager. We included nine masculine characteristics (e.g. dominant, powerful, assertive; α = 0.75) [2] and 10 feminine characteristics (e.g. warm, compassionate, friendly; α = 0.82).

Control variables.

We included as control variables employees’ age (in years), gender (0 = female, 1 = male), educational level (1 = lower than vocational degree, 2 = vocational degree and 3 = university degree), and the percentage of women in management position at their organization (1 = 0%–20%, 2 = 21%–40%, 3 = 41%–60%, 4 = 61%–80% and 5 = 81%–100%), as these might influence employees’ preferences for masculine and feminine leadership qualities (Powell et al., 2021; Stoker et al., 2012).

Results

Descriptive statistics

Table 2 shows the descriptive statistics for the full data set. These results suggest that gender of the respondent was not associated with preferences for masculine leadership characteristics (r = −0.02, p = 0.09) and negatively associated with preferences for feminine leadership characteristics (r = −0.07, p < 0.001). The latter finding suggest that, overall, men rate feminine leadership traits lower than women do. In addition, the perceived percentage of women in management position was not associated with preferences for masculine leadership characteristics (r = −0.02, p = 0.19) and positively associated with preferences for feminine leadership characteristics (r = 0.09, p < 0.001). Age was negatively associated with preferences for masculine leadership characteristics (r = −0.08, p < 0.001) and positively associated with preferences for feminine leadership characteristics (r = 0.03, p = 0.01). Finally, educational level was neither associated with preferences for masculine (r = 0.01, p = 0.69) nor with feminine leadership characteristics (r = −0.00, p = 0.81).

Hypothesis testing

To test H1, that employees value masculine over feminine leadership characteristics, we conducted a one-way (within: masculine vs feminine traits) ANOVA (See row 1 of Table 3). In support of H1, the analysis revealed a main effect of type of leadership trait, F(1,5540) = 3741.86, p < 0.001, Wilk's Λ = 0.60, ηp² = 0.40, such that participants overall favored masculine (M = 4.01, SD = 0.38) over feminine (M = 3.60, SD = 0.44) leadership characteristics.

To test H2, that employees value masculine over feminine leadership characteristics less in the past decade, we conducted ordinary least squares regression analyses with the difference between masculine and feminine preference (higher scores indicate preference for masculine leadership characteristics) as the dependent variable and the two dummy-coded time variables as independent variables (Figure 1 and Table 4). This analysis revealed that, in line with H2, preferences for masculine over feminine leadership decreased in 2020 compared to 2005, B = −0.32, SE = 0.02, p < 0.001, 95% CI [−0.36, −0.28]. The results also showed that the preference for masculine over feminine leadership characteristics increased slightly in 2010 compared 2005, B = 0.05, SE = 0.02, p < 0.001, 95% CI [0.02, 0.08].

To further specify the effect for 2020, we analyzed participants’ preferences over time for masculine and feminine leadership separately. The results showed no difference between 2010 and 2005 with respect to the preference for masculine leadership characteristics, B = −0.002, SE = 0.01, p = 0.87, 95% CI [−0.16, 0.87] and a decreased preference for masculine leadership traits in 2020 compared to 2005, B = −0.12, SE = 0.02, p < 0.001, 95% CI [−0.15, −0.09]. Furthermore, the preference for feminine leadership characteristics decreased between 2005 and 2010, B = −0.06, SE = 0.01, p < 0.001, 95% CI [−0.08, −0.03], while it increased between 2005 and 2020, B = 0.20, SE = 0.02, p < 0.001, 95% CI [0.17, 0.23].

Finally, we note that although the preference for masculine over feminine leadership was lower in 2020 than 2005, the respondents still favored masculine over feminine leadership traits in 2020, F(1,746) = 49.54, Wilk's Λ = 0.94, p < 0.001, ηp² = 0.06 (see final row in Table 3). We further note that for all analyses reported above, including the control variables (i.e. gender, age, educational level and percentage of women in management positions) did not meaningfully change any of the reported results. Moreover, gender did not moderate any of the reported associations.

Matching employees

To further check the robustness of our findings with respect to H2, we performed a matching technique to minimize demographic differences between employees in the 2010 and 2020 samples. Specifically, following Blundell et al. (2004), we matched employees on the variables gender, age, educational level and percentage of women in management positions, using nearest neighbor matching with no caliper (Randolph and Falbe, 2014). We matched our two samples using the MatchIt R package (Ho et al., 2011). We matched the 2010 (N = 1565) sample with the 2020 (N = 748) sample, which resulted in a matched sample of 636 employees. Note that in the 2010 sample, 106 participants indicated they did not know the percentage of women in their organization, 5 participants did not report their age and 1 participant failed to provide an educational level. These participants were excluded from the matched sample. Table 5 presents the demographics of our matched sample. The matching technique was successful, meaning that differences between the two samples on all demographics were reduced, although participants in the 2020 sample were still significantly older and reported a significantly higher percentage of women in management positions.

As in the analysis reported above to test H1, a one-factor (within: masculine vs feminine traits) ANOVA revealed a main effect of the within-factor type of traits, F(1,1270) = 518.34, Wilk's Λ = 0.71, p < 0.001, ηp² = 0.29. In line with H1, employees overall rated masculine leadership traits (M = 3.96, SD = 0.39) higher than feminine leadership traits (M = 3.63, SD = 0.47).

In support of H2, and similar to the results reported above, the analysis showed a negative association between time (coded 0 for 2010 and 1 for 2020) and a preference for masculine over feminine leadership traits, B = −0.44, SE = 0.03, 95% CI [−0.49, −0.39], R²change = 0.19. Separate regression analyses further showed a negative association between time and masculine leadership traits, B = −0.12, SE = 0.02, p < 0.001, 95% CI [−0.16, −0.07], R²change = 0.02, and a positive association between time and preference for feminine leadership traits, B = 0.33, SE = 0.03, p < 0.001, 95% CI [0.27, 0.37], R²change = 0.12.

Finally, we note that for the reported analyses, the inclusion of the control variables as covariates did not meaningfully change the reported results. Moreover, gender did not moderate any of the reported associations.

Discussion

Theoretical and practical implications

The present study examined whether the stereotype of a “good manager” has changed over the last decade. In doing so, we make a substantial contribution to the literature on gendered managerial stereotypes. First, the present study shows that while the preference for masculine over feminine leadership traits slightly increased between 2005 and 2010, over the past decade, we observe a noteworthy decline of employees’ preference for masculine over feminine leadership traits. These results replicate Powell et al.’s (2021) findings, but do so using a sample of working adults instead of MBAs and business students. We show that preferences for masculine leadership traits have decreased, while preferences for feminine leadership traits have grown. In doing so, we demonstrate cross-cultural generalizability and conclude that the stereotype of a “good manager” is changing not only among US business students but also among an older sample of working adults in The Netherlands. Overall, this study increases our confidence that the stereotype of a “good manager” is becoming less gendered.

We further note that the present study did not replicate Powell et al.’s (2021) results regarding the equal importance of masculine and feminine traits (i.e. the androgynous leadership profile). Although we also find that the managerial stereotype has become less masculine and more feminine, the results of our 2020 sample show that employees still prefer masculine over feminine leadership traits. This result might indicate generational differences between Powell et al.’s younger and our older sample, which in turn suggests that in the future, we can also except an equal appreciation of masculine and feminine traits among working adults.

In addition, our results show a slightly increased preference for masculine over feminine leadership traits between 2005 and 2010. This reinforcement of the gendered managerial stereotype could potentially be explained by the financial crisis of that period, as previous research suggests that in times of such an external crisis people look for certainty and thus prefer masculine over feminine leaders (Laustsen and Petersen, 2017; Stoker et al., 2011). Considering these fluctuations in preference for different leadership traits, it is important that we keep tracking potential changes in the managerial stereotypes. Although we observe a rise of feminine leadership over the past decade, this particular finding suggests that the changing managerial stereotype is not necessarily a linear process, but might be reversed or, alternatively, be sped up by external societal circumstances.

The present study also has important practical implications. In fact, changes in the managerial stereotype could prove to be an important step forward for women’s advancement to management and leadership positions. The existence of such stereotypes has disadvantaged women for a long time, as a result of which they are less likely to be hired, and or to apply for, managerial roles (Eagly and Karau, 2002; Heilman, 2001; Schein, 2001). This also creates a greater challenge for women to be seen as effective in leadership roles (Eagly and Karau, 2002). Although other systematic barriers still persist, the fading of the managerial stereotype is a reason for optimism about the future of women in management.

Limitations and future research suggestions

The present study is not without its limitations. Most importantly, although we targeted the same population in all three waves, we used three different samples. Therefore, demographic differences exist between samples across the time periods. These differences could potentially operate as confounds in the reported findings. Although we used a matching technique for the 2010 and 2020 samples, following Blundell et al. (2004), to remove most of the demographic differences between the two samples and although replicating our analyses using these matched samples did not show any meaningful differences, future research would undoubtedly benefit from tracking the same group of individuals over time.

It is also important to note that the BSRI (Bem, 1974) has received substantial criticism over the years, with researchers questioning whether its items adequately represent the constructs of masculinity and femininity (Hoffman and Borders, 2001). We still adopted this measurement, as it is by far the most widely used measurement of feminine and masculine qualities and allowed us to compare our results to previously research (Stoker et al., 2012; Powell and Butterfield, 1979, 1989, 2015; Powell et al., 2002, 2021). Nevertheless, future research could benefit from a more established questionnaire when examining the managerial stereotype.

As noted before, the present study did not replicate Powell et al.’s (2021) findings regarding the existence of an “androgynous” manager profile. If the observed trend of increased preference for feminine and decreased preference for masculine leadership traits continues, one would expect this type of profile to occur in the near future. Hence, future research could benefit from replicating this study once more to see how perceptions of leadership continue to change for respondents in a working context.

Finally, future research should test whether changes in the managerial stereotype indeed improve women’s positions in organizations. Many scholars have pointed out the dysfunctionality of gendered managerial stereotypes for women’s advancement to management positions (Eagly and Karau, 2002; Heilman, 2001; Koenig et al., 2011). For future research, it is important that scholars examine whether a decreased preference for masculine and increased preference for feminine leadership traits indeed translate into better opportunities for female managers and an actual increase in their numbers.

Figures

Figure 1.

Preference for masculine and feminine leadership characteristics over time

Table 1.

Demographics of the three samples

Data collection	N	% of women	Age (in years)	Educational level	% of women in management
Sample 2005	3229	33	35.89 (6.17)	2.50 (0.50)	1.40 (0.76)
Sample 2010	1565	48	38.59 (8.34)	2.52 (0.55)	1.71 (1.04)
Sample 2020	747	57	47.41 (10.64)	2.44 (0.63)	2.41 (1.21)

Notes:

Education level was coded as 1 = lower than vocational degree, 2 = vocational degree, and 3 = university degree or higher. Percentage of women in management positions was coded as: 1 = 0%–20%, 2 = 21%–40%, 3 = 41%–60%, 4 = 61%–80% and 5 = 81%–100%. The data collected in 2005 are also reported in Stoker et al., 2012

Table 2.

Descriptive statistics

Variables	M (SD)	1	2	3	4	5
1. Age	38.20 (8.47)	–	–	–	–	–
2. Gender	0.59 (0.49)	0.11***	–	–	–	–
3. Educational level	2.50 (0.53)	−0.09***	−0.08***	–	–	–
4. % of women in management	1.61 (0.97)	0.12***	−0.26***	0.03*	–	–
5. Masculine characteristics	4.01 (0.38)	−0.08***	−0.02	0.01	−0.02	–
6. Feminine characteristics	3.60 (0.44)	0.03*	−0.07***	−0.00	0.09***	–

Notes:

For gender 0 = female, 1 = male, 2 = other. For education level, 1 = lower than vocational degree, 2 = vocational degree, and 3 = university degree. For women in management, 1 = 0%–20%, 2 = 21%–40%, 3 = 41%–60%, 4 = 61%–80% and 5 = 81%–100. *p < 0.05, **p < 0.01, ***p < 0.001. The data collected in 2005 are also reported in Stoker et al., 2012

Table 3.

Mean levels (and standard deviations) of masculine and feminine characteristics over time

	Masculine characteristics		Feminine characteristics
Data collection	M	SD	M	SD	F	df	Wilk's Λ	ηp2
All samples combined	4.01	0.38	3.60	0.44	3741.86***	1,5540	0.60	0.40
2005	4.03	0.37	3.59	0.44	2501.72***	1,3228	0.56	0.44
2010	4.03	0.39	3.53	0.44	1683.67***	1,1564	0.48	0.52
2020	3.90	0.36	3.79	0.40	49.54***	1,746	0.94	0.06

Notes:

Each row reflects a one-way (within: masculine vs feminine characteristics) ANOVA. ***p < 0.001, The data collected in 2005 are also reported in Stoker et al., 2012

Table 4.

Regression results

	Masculine–feminine characteristics		Masculine characteristics		Feminine characteristics
	B (SE)	95% CI	B (SE)	95% CI	B (SE)	95% CI
Constant	0.44 (0.01)***	[0.42, 0.45]	4.03 (0.01)***	[4.01, 4.04]	3.59 (0.01)***	[3.58, 3.61]
Dummy 2010	0.05 (0.02)***	[0.02, 0.08]	−0.00 (0.01)	[−0.03, 0.02]	−0.06 (0.01)***	[−0.08, −0.03]
Dummy 2020	−0.32 (0.02)***	[−0.36, −0.28]	−0.12 (0.02)***	[−0.15, −0.09]	0.20 (0.02)***	[0.17, 0.23]
R2	0.06***		0.01***		0.03***

Notes:

For dummy 2010, 0 = 2005/2020 and 1 = 2010. For dummy 2020, 0 = 2005/2010, 1 = 2020. ***p < 0.001, The data collected in 2005 are also reported in Stoker et al., 2012

Table 5.

Demographics of the matched samples

Data collection	N	% of women	Age (in years)	Educational level	% of women in management
Sample 2010	636	53	43.62 (8.97)	2.48 (0.59)	2.19 (1.28)
Sample 2020	636	58	47.06 (10.57)	2.44 (0.64)	2.40 (1.21)

Notes:

Education level was coded as 1 = lower than vocational degree, 2 = vocational degree, and 3 = university degree or higher. Percentage of women in management positions was coded as: 1 = 0%–20%, 2 = 21%–40%, 3 = 41%–60%, 4 = 61%–80% and 5 = 81%–100%

Notes

1. One participant selected “other” for gender.

2. In line with Stoker et al. (2012), we did not include the item “aggression” for the masculinity scale.