Significance
We identify a profound and consistent gender gap in people’s core life goals. Across nine studies using diverse sample populations (executives in high-power positions, recent graduates of a top MBA program, undergraduate students, and online panels of working adults) and over 4,000 participants, we find that, compared to men, women have a higher number of life goals, place less importance on power-related goals, associate more negative outcomes (e.g., time constraints and tradeoffs) with high-power positions, perceive power as less desirable, and are less likely to take advantage of opportunities for professional advancement. Women view high-level positions as equally attainable as men do, but less desirable. Our findings advance the science of gender, goals, organizational behavior, and decision making.
Keywords: gender, professional advancement, goals, power, achievement
Abstract
Women are underrepresented in most high-level positions in organizations. Though a great deal of research has provided evidence that bias and discrimination give rise to and perpetuate this gender disparity, in the current research we explore another explanation: men and women view professional advancement differently, and their views affect their decisions to climb the corporate ladder (or not). In studies 1 and 2, when asked to list their core goals in life, women listed more life goals overall than men, and a smaller proportion of their goals related to achieving power at work. In studies 3 and 4, compared to men, women viewed high-level positions as less desirable yet equally attainable. In studies 5–7, when faced with the possibility of receiving a promotion at their current place of employment or obtaining a high-power position after graduating from college, women and men anticipated similar levels of positive outcomes (e.g., prestige and money), but women anticipated more negative outcomes (e.g., conflict and tradeoffs). In these studies, women associated high-level positions with conflict, which explained the relationship between gender and the desirability of professional advancement. Finally, in studies 8 and 9, men and women alike rated power as one of the main consequences of professional advancement. Our findings reveal that men and women have different perceptions of what the experience of holding a high-level position will be like, with meaningful implications for the perpetuation of the gender disparity that exists at the top of organizational hierarchies.
Even in societies that stress the importance of gender equality, women are underrepresented in most senior-level positions (1). For instance, recent estimates indicate that women comprise less than 5% of Fortune 500 CEOs, less than 15% of executive officers, less than 20% of full professors in the natural sciences, and only 6% of partners in venture capital firms (2–4). Moreover, it seems that the small percentage of women who do attain executive positions are relegated to spheres within the organization that have less influence and fewer opportunities for professional advancement (5). These differences may not exist at the start of the employment relationship. In fact, recent evidence has found that female applicants were favored over male ones for positions as assistant professors in science (6).
Many reasons exist for the gender imbalance in high-level positions. These reasons can be grouped into two broad categories. The first category is what sociologists refer to as demand-side factors and psychologists call interpersonal effects. These factors encompass the institutional barriers that women face because of the divergent ways in which men and women are perceived and treated by others. For example, evidence suggests that compared to men, women are perceived as less competent and lacking leadership potential (7–9) and are more likely to encounter challenges, skepticism, and backlash about their ideas and abilities (10–15). Interestingly, gender-based biases and discrimination seem to be perpetuated by men and women alike (16, 17).
The second category is what sociologists refer to as supply-side factors and psychologists call intrapersonal effects. In contrast to demand-side factors, which are part of the environment the individual interacts with, supply-side factors are differences in the perceptions held, decisions made, or behaviors enacted by men and women themselves that contribute to gendered outcomes. For example, men are more likely than women to engage in dominant or aggressive behaviors (18–22), to initiate negotiations (23), and to self-select into competitive environments (24–26)—behaviors likely to facilitate professional advancement.
In this paper, we examine a supply-side factor that has received little research attention: male vs. female preferences for achieving high-level positions in the workplace. Specifically, we focus on people’s life goals and the positive and negative outcomes men and women associate with professional advancement.
The goals people set for themselves are a powerful motivator of their current behavior (27). In addition to being driven by their beliefs about what will make them happy in life, people’s goals are determined by the way they imagine their future to be (e.g., having a certain job or a specific set of relationships). The images people hold about the future are affected by sociocultural factors (e.g., family values or attitudes toward work) as well as norms and expectations that define the context in which they live (28). In recent decades, women’s roles have changed more dramatically than those of men, at least in Western societies and cultures (28). Although women are still interested in pursuing goals related to having strong relationships, marriage, and family, they are also increasingly interested in being professionally employed and having a career. As a result, we hypothesize that women are likely to have more life goals than men, reflecting a greater diversity of preferences for what they hope to accomplish in the future.
In addition to a difference in total number of goals, we also predict that, compared to men’s life goals, a smaller proportion of women’s life goals are related to achieving power at work. Previous research provides some evidence that men are more likely to strive for power in the workplace than are women. Men tend to pay more attention to power cues (29) and to be more motivated by power—the desire for the means to influence other people (30)—which has been shown to play a role in producing gender differences in leadership role occupancy (31). In contrast, women tend to be more motivated by affiliation—the desire for warm, close relationships with others (32). Reinforcing this point, in studies using samples of talented individuals, the life values and personal views of men and women have been found to differ (33), with men tending to assume a more agentic, career-focused perspective, whereas women generally favor a more communal, holistic perspective (34–36). These differing views seem to cause differences in how men and women allocate their time and attention (33).
Finally, we expect that women associate power-related goals (e.g., taking on a high-level position in an organization) with more negative outcomes than men do because pursuing a power-related goal is more likely to conflict with their other life goals. Compared to women, men perceive a longer time frame for achieving their main life goals (28). By perceiving a subjectively shorter time frame within which they must attain their goals and by having a greater diversity of goals, women are likely to experience more conflict among their goals. When one of their goals is salient and its attainability is close (e.g., they are offered a promotion at work), women are more likely than men to feel anxious due to the sacrifices or difficult tradeoffs they would have to make to give that one goal more attention than others. As a result, compared to men, we expect women to view a high-level position as less desirable, even if it seems equally attainable.
To shed light on these hypotheses, we conducted nine studies that use a variety of methods and sample populations. Across our studies, we seek to capture women’s and men’s preferences for professional advancement and power in the workplace, independent of the causes of those preferences.
Results
Studies 1 and 2: Gender and Life Goals.
Our first hypotheses are that (i) women have more life goals than men do overall and (ii) a smaller percentage of those goals are related to power. To test these predictions, in studies 1 and 2 we asked participants to list their core life goals.
In study 1, we asked a large online sample of working adults (n = 781) to write a list of their core goals in life. We defined core goals for participants as “things that occupy your thoughts on a routine basis, things that you deeply care about, or things that motivate your behavior and decisions. Examples of such goals are: being in a committed relationship, keeping up with sports, being organized, or attaining power or status.” This description was based on previous research on personal strivings, defined as consciously accessible and personally meaningful objectives that people pursue in their daily lives (37–39). We asked participants to list anywhere from 1 to 25 goals (in the order in which they came to mind) within two minutes. Participants then categorized their goals by choosing among different goal categories with descriptions (Supporting Information), which were based on research on personal strivings (37–39).
As expected, women listed more goals than did men [meanf = 9.46, SDf = 5.63 vs. meanm = 8.41, SDm = 5.28, t(779) = 2.67, P = 0.008, d = 0.19] and listed a smaller proportion of power-related goals out of their list of total goals [meanf = 3%, SDf = 9% vs. meanm = 7%, SDm = 14%, t(779) = 4.18, P < 0.001, d = 0.34]. There was also a main effect of gender on miscellaneous goals such that women reported a higher proportion of such goals than did men [meanf = 12%, SDf = 17% vs. meanm = 9%, SDm = 16%, t(779) = 2.60, P = 0.01, d = 0.18]. We found no gender differences among the other goal dimensions: achievement, affiliation, personal growth and health, generativity, spirituality, or avoidance.
In study 2, we provide a conceptual replication of study 1 and also address a potential confound: that women listed more goals than men because they cared more about pleasing the experimenter. We asked 437 adults from an online panel of employed individuals provided by ClearVoice to complete a short survey. Participants listed their core goals (this time, up to 20 of them) and then categorized them, using the same categories as in study 1. After listing their goals, to test the alternative explanation about level of effort in the study, we asked participants to list their favorite foods under the same two minute time limit. Finally, after answering demographic questions, participants indicated the extent to which, while completing the study, they tried to please the experimenter on a 7-point scale (1 = not at all, 7 = very much so).
Mirroring the results of study 1, compared to men, women listed more goals [meanf = 9.47, SDf = 4.81 vs. meanm = 7.90, SDm = 4.63, t(435) = 3.48, P = 0.001, d = 0.34]. Importantly, female and male participants listed about the same number of favorite foods [meanf = 13.54, SDf = 5.21 vs. meanm = 12.76, SDm = 5.70, t(435) = 1.50, P = 0.14, d = 0.14] and were similarly disinterested in pleasing the experimenter during the study [meanf = 2.69, SDf = 2.06 vs. meanm = 2.68, SDm = 1.91, t(435) < 1, P = 0.96, d = 0.005]. Even when controlling for the number of favorite foods respondents reported, women still listed more life goals than men, F(1,434) = 10.23, P = 0.001, = 0.023.
Female participants again listed a smaller proportion of power-related goals out of their list of total goals than did men [meanf = 5%, SDf = 11% vs. meanm = 10%, SDm = 13%, t(435) = 4.51, P < 0.001, d = 0.42]. There was also a main effect of gender on avoidance goals such that women reported a higher proportion of avoidance goals than did men [meanf = 2%, SDf = 7% vs. meanm = 1%, SDm = 4%, t(435) = 2.15, P = 0.032, d = 0.18]. We found no gender differences among the other goal dimensions.
Studies 3 and 4: Desirability vs. Attainability of Professional Advancement.
We predicted that having more goals overall and a small percentage of them related to power would lead women to see opportunities for professional advancement as less desirable than men do, but equally attainable. We test these predictions in studies 3 and 4.
In study 3, we contacted 1,762 MBA students who had graduated from a top MBA program in the last two years (i.e., 2013 and 2014) via email and asked them to fill out a short survey. Six hundred thirty-five of them replied. Participants were shown a ladder with rungs numbered from 1 to 10 and told to imagine it represented the hierarchy of professional advancement in their current professional industry. We asked them to think about their career and to indicate three different positions (i.e., rungs) on the ladder: (i) their current position in their industry, (ii) their ideal position, and (iii) the highest position they could realistically attain.
There were no significant differences between men and women in the current position that they reported [meanf = 5.39, SDf = 1.85 vs. meanm = 5.63, SDm = 2.11, t(633) = −1.46, P = 0.145, d = 0.12]. Controlling for their current position, compared to male participants, female participants reported a lower ideal position [meanf = 9.04, SDf = 1.14 vs. meanm = 9.59, SDm = 0.92, F(1,632) = 41.99, P < 0.001, = 0.062]. However, the highest attainable positions reported by men and women were equally high [meanf = 9.29, SDf = 0.88 vs. meanm = 9.41, SDm = 0.98, F(1,632) = 1.58, P = 0.21, = 0.002].
In study 4, we conceptually replicated the findings from study 3 by investigating people’s perceived desirability and attainability of professional advancement. We asked an online panel of 247 adults to “think about the things you’d like to accomplish in your life and the goals you have for yourself.” We specifically asked participants to think about two statements that focused on goals related to professional advancement: (i) “As one of my core goals in life, I would like to have a powerful position in an organization,” and (ii) “As one of my core goals in life, I would like to have power over others.” Participants rated the desirability and attainability of these two goals on 7-point scales (1 = not at all, 7 = very much so).
Compared to male participants, female participants rated these goals as less desirable [meanf = 3.73, SDf = 1.67 vs. meanm = 4.19, SDm = 1.52, t(245) = 2.24, P = 0.026, d = 0.29], but equally attainable [meanf = 4.25, SDf = 1.67 vs. meanm = 4.35, SDm = 1.56, t(245) = 0.49, P = 0.63, d = 0.06].
Studies 5–7: Positive and Negative Outcomes Associated with Professional Advancement.
In studies 5–7, we explore why women see professional advancement as less desirable by asking people to rate the likelihood of experiencing positive outcomes (e.g., opportunity, satisfaction) and negative outcomes (e.g., goal conflict, time constraints) upon receiving a promotion at work.
In study 5, we asked 465 working adults from an online database to imagine being promoted to a higher-level position in their current organization. Participants were told that as a result of this promotion, their level of power over others would increase substantially. Participants predicted the extent to which they would experience nine different outcomes if they received the promotion (on a 10-point scale). Some outcomes were positive (satisfaction or happiness, opportunity, money, status, or influence) whereas others were negative (stress or anxiety, difficult tradeoffs or sacrifice, time constraints, burden of responsibility, or conflict with other life goals). Participants also indicated how desirable the promotion would be to them and their likelihood of pursuing the promotion (on 7-point scales).
Participants’ ratings on the nine outcomes loaded onto two separate factors: positive outcomes (α = 0.77) and negative outcomes (α = 0.84). Compared to male participants, female participants expected stronger negative outcomes to occur with the promotion [meanf = 6.73 out of 10, SDf = 1.56 vs. meanm = 6.23, SDm = 1.39, t(442) = 3.43, P = 0.001, d = 0.34]. However, men and women expected a statistically equivalent level of positive outcomes to occur with the promotion [meanf = 7.13, SDf = 1.44 vs. meanm = 7.23, SDm = 1.28, t(442) = 0.79, P = 0.43, d = 0.07]. Female participants also viewed the potential promotion as less desirable than men did [meanf = 5.12, SDf = 1.46 vs. meanm = 5.48, SDm = 1.32, t(442) = 2.6, P = 0.01, d = 0.26] and indicated that they would be less likely to go after the promotion [meanf = 4.77, SDf = 1.41 vs. meanm = 5.25, SDm = 1.43, t(442) = 3.35, P = 0.001, d = 0.34]; see Fig. 1.
Fig. 1.
Ratings of desirability of receiving a promotion or obtaining a high-power position by gender in studies 5–7. Error bars represent SEs. *P ≤ 0.05, **P ≤ 0.01.
We also found that the negative outcomes expected by women explained (i) women's perception of the promotion as less desirable and (ii) their reduced likelihood of pursuing the promotion. When including expected negative outcomes as a mediator and expected positive outcomes as a control variable, the effect of gender on desirability of the promotion weakened (from β = 0.10, P = 0.008 to β = 0.046, P = 0.19), and expected negative outcomes predicted lower desirability of the promotion (β = −0.33, P < 0.001; 95% bias-corrected confidence interval (CI) = [0.07, 0.26]). Similar regression analyses using likelihood of pursuing the promotion as the dependent measure revealed a weakened effect of gender (from β = 0.14, P = 0.001 to β = 0.098, P = 0.011) and a negative effect of negative outcomes associated with the potential promotion (β = −0.24, P < 0.001; 95% bias-corrected CI for the size of the indirect effect = [0.05, 0.21]). We found similar results (Supporting Information) when conducting mediation analyses focused on the single item “conflict with other life goals” or an aggregate of the two items “conflict with other life goals” and “difficult tradeoffs.”
In study 6, we tested the same relationships in a sample of executives who were likely to already occupy positions of power and had displayed an interest in furthering their careers by enrolling in executive education courses focused on leadership, decision making, and negotiation at a top US business school. Two hundred and four executives completed the same scenario and measures as in study 5 as part of their required coursework.
Mirroring the results of study 5, compared to male participants, female participants associated more negative outcomes with the hypothetical promotion [meanf = 7.09 out of 10, SDf = 1.71 vs. meanm = 6.17, SDm = 2.33, t(202) = 2.61, P = 0.01, d = 0.45], although men and women anticipated statistically equivalent levels of positive outcomes being associated with the promotion [meanf = 7.39, SDf = 1.38 vs. meanm = 7.11, SDm = 1.43, t(202) = 1.22, P = 0.22, d = 0.20]. Female participants also reported viewing the potential promotion as less desirable [meanf = 5.69, SDf = 1.15 vs. meanm = 6.02, SDm = 0.91, t(202) = 2.09, P = 0.038, d = 0.32] and indicated that they would be less likely to pursue the promotion compared to male participants [meanf = 5.71, SDf = 1.33 vs. meanm = 6.07, SDm = 1.15, t(202) = 1.84, P = 0.067, d = 0.29], as depicted in Fig. 1.
Also, as in study 5, we found that the negative outcomes women believed they would experience after achieving a higher-level position explained their perception of the promotion as less desirable and their reduced likelihood of pursuing the promotion. Controlling for expected positive outcomes, the effect of gender on desirability of the promotion weakened (from β = 0.17, P = 0.01 to β = 0.13, P = 0.044), and negative outcomes associated with the promotion predicted lower desirability of the promotion (β = −0.28, P < 0.001; 95% bias-corrected CI = [0.01, 0.26]). Similar regression analyses using likelihood of pursuing the promotion as the dependent measure revealed a weakened effect of gender (from β = 0.15, P = 0.029 to β = 0.109, P = 0.104) and a negative effect of negative outcomes associated with the potential promotion (β = −0.27, P < 0.001; 95% bias-corrected CI = [0.01, 0.31]). We also find similar results (Supporting Information) when conducting mediation analyses focused on the single item “conflict with other life goals” or an aggregate of the two items “conflict with other life goals” and “difficult tradeoffs.”
In study 7, we slightly modified the design of study 6 to examine whether the results held in a sample of undergraduates at a top US university. Similar to the executive sample, these individuals were likely to be interested in professional advancement based on the prestigious academic environment they had elected to be a part of; however, in contrast to the executive sample, these participants had not yet entered the professional workforce.
We asked 516 undergraduate students to imagine that upon graduation they were presented with a high-power job opportunity. We asked them to describe the high-power job they were imagining and then list the outcomes or feelings they would associate with occupying such a position. Participants could list anywhere from 1 to 15 items. Participants also indicated how desirable the position would be to them, their likelihood of pursuing the position if it required extra effort, and their likelihood of pursuing the position if it did not require any extra effort (on 7-point scales). Next, we presented participants with the outcomes they anticipated experiencing with the high-power job and asked them to categorize these outcomes as positive, neutral, or negative.
Compared to male participants, female participants spontaneously listed a lower proportion of positive outcomes [meanf = 50.1%, SDf = 38.9% vs. meanm = 56.6%, SDm = 32.9%, t(514) = 2.05, P = 0.041, d = 0.18] and a higher proportion of negative outcomes [meanf = 30%, SDf = 37% vs. meanm = 22.5%, SDm = 25.8%, t(514) = 2.65, P = 0.008, d = 0.24]. The proportion of neutral outcomes listed was the same for men and women [meanf = 19.9%, SDf = 29.9% vs. meanm = 20.9%, SDm = 24.2%, t(514) = 0.39, P = 0.70, d = 0.03].
As depicted in Fig. 1, female participants reported viewing the high-power position as less desirable than male participants [meanf = 5.02, SDf = 1.09 vs. meanm = 5.37, SDm = 1.43, t(514) = 3.16, P = 0.002, d = 0.28]. Compared to men, women were also less likely to pursue the position, regardless of whether it necessitated extra effort on their part [meanf = 5.07, SDf = 1.14 vs. meanm = 5.41, SDm = 1.41, t(514) = 3.09, P = 0.002, d = 0.27] or not [meanf = 5.54, SDf = 1.09 vs. meanm = 5.97, SDm = 1.15, t(514) = 4.29, P < 0.001, d = 0.38].
Studies 8 and 9: Professional Advancement and Power.
In studies 5–7, the potential promotion or high-level position was described as being high in power. In studies 8 and 9, we manipulated the description of the high-level position to examine whether the effects in our previous studies hinged upon the emphasis placed on power.
In study 8, 484 online participants were asked to imagine the possibility of being promoted to a higher-level position in their current organization, similar to studies 5 and 6. Unlike in previous studies, participants were randomly assigned to one of two promotion description conditions: (i) promotion with power or (ii) promotion with power defined. In the promotion with power condition, participants were told, “As a result of this promotion, your level of power would increase substantially” (as participants were told in studies 5 and 6). Participants in the promotion with power defined condition were additionally told, “By power, we mean your relative ability to control other people’s outcomes, experiences, or behaviors.”
We then asked participants to indicate the extent to which they thought such a promotion would conflict with their other life goals and to what extent it would require them to make tradeoffs and sacrifices (on 7-point scales). We aggregated across these two items to create a measure of expected conflict (α = 0.85). Participants then rated the desirability of the promotion (on a 7-point scale) and answered the question “How do you view having power in a job?” by selecting one of two possible answers: (i) “This is a goal that I am not that interested in pursuing” or (ii) “This is a goal I definitely want to pursue.” Finally, we asked participants to list two or three words they associated with having power at work.
The promotion description manipulation (promotion with power vs. promotion with power defined) did not have a significant effect on expected conflict [F(1,480) = 1.44, P = 0.23, = 0.003] and had only a marginally significant effect on desirability of the promotion [F(1,480) = 3.13, P = 0.08, = 0.005]. The interactions between the promotion description manipulation and gender were insignificant for both expected conflict [F(1,480) = 1.15, P = 0.28, = 0.002] and promotion desirability [F(1,480) < 1, P = 0.98, = 0.000].
Compared to male participants, female participants rated the promotion as less desirable [meanf = 5.18, SDf = 1.50 vs. meanm = 5.45, SDm = 1.32, F(1,480) = 4.36, P = 0.037] and anticipated experiencing more conflict with other life goals [meanf = 3.81, SDf = 1.58 vs. meanm = 3.41, SDm = 1.37, F(1,480) = 8.26, P = 0.004]. Additionally, more female than male participants [41.0% (75/183) vs. 30.9% (93/301)] indicated that they were not interested in pursuing power as a goal [χ2(1,N = 484) = 5.11, P = 0.024, Cramér’s V = 0.10].
We predicted that the conflict women expect to experience in a position of high power explains their lower ratings of promotion desirability. Thus, we conducted regression analyses with desirability of the promotion as the dependent measure, expected conflict with other goals as the potential mediator, and the promotion description manipulation as the control variable. The effect of gender weakened (from β = 0.095, P = 0.037 to β = 0.031, P = 0.44) and expected conflict with other goals predicted lower ratings of promotion desirability (β = −0.49, P < 0.001; 95% bias-corrected CI = [0.06, 0.32]).
The words participants associated with having power at work were coded by four gender-blind coders, two female and two male. Each word was assigned a code of positive, negative, or neutral, and we averaged across the codes provided by all four coders for our analyses (all α >0.70, average α = 0.83). Mirroring the pattern of results found in our other studies, we found no gender differences in the number of positive words [meanf = 1.49, SDf = 0.90 vs. meanm = 1.52, SDm = 0.79, F(1,483) = 0.24, P = 0.627, = 0.000] or neutral words [meanf = 0.79, SDf = 0.57 vs. meanm = 0.83, SDm = 0.57, F(1,483) = 0.68, P = 0.41, = 0.001] listed by participants, but female participants listed significantly more negative words associated with having power at work than men did [meanf = 0.50, SDf = 0.73 vs. meanm = 0.37, SDm = 0.65, F(1,483) = 4.56, P = 0.033, = 0.009].
In study 9, we sought to replicate the results of study 8 in a sample of executives, and we added an experimental condition that completely removed all explicit emphasis on power as a necessary outcome of professional advancement. We collected data from 265 executives enrolled in executive education courses focused on influence, decision making, and negotiation at a top US business school. We assigned participants to one of three promotion description conditions: (i) promotion with power, (ii) promotion with power defined, or (iii) basic promotion. The prompts for the first two conditions were the same as in study 8. In the basic promotion condition, an increase in power was not mentioned as a result of the promotion.
Participants were then presented with the same nine outcomes as in studies 5–7 and asked to report how much they expected to experience each of them if they received the promotion. They also indicated the desirability of the promotion and their likelihood of pursuing it.
Similar to the findings of the previous study, the promotion description did not affect the results. The results of 2 (men vs. women) × 3 (promotion with power vs. promotion with power defined vs. basic promotion) ANOVAs conducted on dependent measures revealed no main effects for the promotion description manipulation or significant interactions (all P > 0.11). Across our analyses, the only significant effects were main effects of gender on our dependent measures. Table 1 reports the means and SDs of the main variables we measured by gender across conditions.
Table 1.
Means (and SDs) of the main variables measured in study 9 by gender across conditions
Promotion with power | Promotion with power defined | Basic promotion | ||||
Variable | Male | Female | Male | Female | Male | Female |
Negative outcomes | 5.19 (1.35) | 5.87 (0.96) | 5.06 (1.33) | 5.72 (1.12) | 4.80 (1.69) | 5.43 (1.27) |
Positive outcomes | 6.15 (0.63) | 6.01 (0.59) | 6.38 (0.68) | 6.29 (0.94) | 6.25 (1.08) | 6.11 (1.03) |
Desirability of the promotion | 6.13 (1.09) | 5.23 (1.50) | 6.08 (1.11) | 5.42 (1.44) | 5.94 (1.18) | 5.37(1.22) |
Likelihood of pursuing a promotion | 6.19 (1.23) | 4.94 (1.71) | 5.65 (1.30) | 5.07 (1.56) | 5.63 (1.50) | 4.91 (1.25) |
Compared to male participants, female participants associated more negative outcomes with the promotion [meanf = 5.68, SDf = 1.13 vs. meanm = 5.02, SDm = 1.46; F(1,259) = 15.89, P < 0.001, = 0.06] but about the same level of positive outcomes [meanf = 6.15, SDf = 0.88 vs. meanm = 6.26, SDm = 0.82; F(1,259) = 1.36, P = 0.25, = 0.005]. Women also reported viewing the potential promotion as less desirable than did men [meanf = 5.35, SDf = 1.38 vs. meanm = 6.05, SDm = 1.12; F(1,259) = 20.90, P < 0.00, = 0.075] and indicated that they would be less likely to pursue it [meanf = 4.98, SDf = 1.51 vs. meanm = 5.83, SDm = 1.36; F(1,259) = 22.93, P < 0.001, = 0.81].
Taken together, the results of studies 8 and 9 replicate the results of studies 5–7 and suggest that our findings are not contingent on the emphasis placed on power as a corollary of professional advancement. In light of previous research showing differences in power orientation between men and women, it is unlikely that these results mean that power does not play an important role in the gender differences we observe. Rather, we suspect that people automatically associate professional advancement with an increase in power, and reinforcing this association is merely redundant, yielding no additional effect.
Discussion
Across nine studies using diverse sample populations (executives in high-power positions, graduates of a top MBA program, undergraduate students, and online panels of working adults) and over 4,000 participants, we find that, compared to men, women have a higher number of life goals, place less importance on power-related goals, associate more negative outcomes (e.g., goal conflict and tradeoffs) with high-power positions, perceive power as less desirable though equally attainable, and are less likely to take advantage of opportunities for professional advancement. In our research, we used a definition of power (i.e., the desire for the means to influence other people) that has been commonly used in extant literature. Definitions that encompass other types of power (e.g., helping the organization achieve its objectives or run more effectively) may lead to different results, and thus exploring their potentially unique impact may be a fruitful avenue for future work.
Identifying the origin of the differences between men’s and women’s professional aspirations is beyond the scope of the current research. Our findings may be the result of biological gender differences, learned preferences that have developed in response to cultural norms and gender-based discrimination, or both. In addition, supply-side factors (e.g., personal goals) and demand-side factors (e.g., gender-based backlash and discrimination) are inextricably linked. People learn how to think and behave based on their experiences, observations, and interactions in the world. For example, work on gender and volubility has shown that women speak up less often than do men owing to an acute awareness of the backlash that women frequently receive for voicing their opinions (40). Similarly, a woman may innately desire power, but she may see how women in high-level positions act and are treated and decide that power is an undesirable goal for her.
It is also important to note that our findings are descriptive, not prescriptive. Based on these data, we cannot make value judgments about whether men’s and women’s differing views of professional advancement are good or bad, rational or irrational, at any level of analysis (e.g., for individuals, for organizations, or for societies). It is possible that men and women are correctly predicting the unique experiences that they are poised to encounter upon professional advancement and are making sound decisions accordingly (41–43). For instance, if women who hold the same positions as men at work are required to complete more tasks outside of work for themselves and/or their families, then the differences we observe may be optimal (44). However, it is also possible that women are overestimating the negative consequences associated with power, that men are underestimating them, or both. Future work could explore the congruence between the predicted and actual experience of achieving high-power positions for men and women.
Scholars have considered the possibility that supply-side impediments prevent women from achieving power and status at work, whereas other research has found evidence that women face demand-side barriers in the workplace. Our findings break new ground by documenting a previously unstudied supply-side phenomenon: compared to men, women have more life goals that make achieving high-power positions at work seem less desirable (but equally attainable). Therefore, women may not assume high-level positions in organizations—at least in part—because they desire other things as well.
Materials and Methods
Here we describe the sample populations we recruited in our nine studies. For additional methodological detail, full results, and tables, refer to Supporting Information. We obtained informed consent from all participants, and the Institutional Review Board of Harvard University reviewed and approved all materials and procedures in our experiments.
Study 1.
We recruited 800 adult participants on Amazon Mechanical Turk to participate in a study for $1. Only participants located in the United States were allowed to complete the study.
We calculated our sample size based on an estimated effect size d = 0.2, requiring a sample size of ∼790 participants for a study powered at 80%. Nineteen participants did not complete the survey; our final sample size was 781 participants (meanage = 31.88, SD = 8.40, 58% female).
Study 2.
We recruited 437 adults (meanage = 48.41, SD = 11.50, 50% female) from ClearVoice, an online panel of employed individuals provided by an organization working with academic institutions, to complete a short survey in exchange for $1. When we contacted ClearVoice, we asked for 500 respondents, 50% male and 50% female, all employed, knowing that we might not be able to obtain the full sample by the study deadline. We only obtained complete data from 437 respondents before the study deadline. We calculated our sample size based on an estimated effect size d = 0.3, requiring a sample size of ∼470 participants for a study powered at 90%.
Study 3.
We contacted 1,762 MBA students who had graduated from a top MBA program in the United States in the last two years (i.e., 2013 and 2014) via email and asked them to fill out a short survey by a given deadline. Six hundred thirty-five (meanage = 29.93, SD = 2.14, 39% female) replied by the deadline, corresponding to a 36% response rate. Respondents were not paid for their participation.
We calculated our sample size based on an estimated effect size d = 0.3, requiring a sample size of ∼580 participants for a study powered at 95%. We expected a response rate of about 30%, when in fact it was higher.
Study 4.
We recruited an online panel of 247 adults from Amazon Mechanical Turk (meanage = 29.70, SD = 9.51, 44.9% female) to participate in a study in exchange for $1. Only participants located in the United States were allowed to complete the study. We recruited 250 participants; three of them did not complete the survey, so our final sample size was 247 participants.
Study 5.
Four hundred sixty-five working adults (meanage = 29.46, SD = 7.83, 35% female) from Amazon Mechanical Turk participated in the study in exchange for $0.50. We calculated our sample size based on an estimated effect size d = 0.3, requiring a sample size of ∼470 participants for a study powered at 90%.
Study 6.
Participants for this study were enrolled in executive education courses that focused on leadership, decision making, and negotiation at a top US business school. They were all mid- to senior-level managers from a broad cross-section of industries, and they were not compensated for their participation. Two hundred four people (meanage = 37.94, SD = 6.30, 52 female, 152 male) completed the study as part of their classwork.
Study 7.
Five hundred thirty undergraduates at a top university in the United States participated in exchange for a $10 Amazon gift card (meanage = 22.14, SD = 3.15, 46.9% male).
We calculated our sample size based on an estimated effect size d = 0.3, requiring a sample size of ∼580 participants for a study powered at 95%. However, we obtained only 530 responses by the deadline we imposed before running the study.
Study 8.
Four-hundred eighty-four people (meanage = 32.56, SD = 9.54, 60.2% male) from Amazon Mechanical Turk participated in the study for $0.50. We calculated our sample size based on an estimated effect size f = 0.15, requiring a sample size of ∼490 participants for a study powered at 80%.
Study 9.
We collected data from 265 executives enrolled in executive education courses focusing on influence, decision making, and negotiation at a top US business school. Study participation was completed as part of their required coursework.
Study 1
Method.
Participants first read initial instructions welcoming them to the study and answered an attention check. Those who failed the attention check were automatically taken to a screen letting them know that, based on their answers, they did not qualify for the study. Thus, their data were not recorded.
Next, we asked participants to write a list of the goals that they would like to accomplish in their lives or that they considered to be of central importance to themselves. We defined core goals as “things that occupy your thoughts on a routine basis, things that you deeply care about, and/or things that motivate your behavior, decisions, etc.” This description was based on prior conceptualizations of personal strivings—consciously accessible and personally meaningful objectives that people pursue in their daily lives (1, 2). We provided participants with some examples: “being in a committed relationship,” “keeping up with sports,” “being organized,” or “attaining power or status.” We gave them 25 open text boxes and asked them to write each goal in a box. They could list anywhere from 1 to 25 goals; we did not ask them to rank the goals in order of importance but simply to list them in the order in which they came to mind. They were given 2 min to list their goals.
Next, we presented participants with the goals they had listed and asked them to indicate the category to which the goals belonged using a set of categories we provided: power, achievement, affiliation, personal growth and health, generativity, spirituality, avoidance, and miscellaneous. For each of these categories, we included a brief description (Table S1). These goal categories were based on the goals listed by previous research on personal strivings (1, 2). Finally, participants reported their age, gender, and number of children.
Table S1.
Goal category descriptions provided to participants in studies 1 and 2
Goal category | Description |
Power | To have an impact on, control or manage other people, influence other people, or control resources others depend on |
Achievement | To achieve success by doing things better than before or by surpassing some internal or external standards of excellence |
Affiliation | To have close and warm relationships with others (e.g., spend time with others, be involved in social relationships) |
Personal growth and health | To engage in behaviors that improve one’s health or knowledge (e.g., to eat more vegetables, get in better shape, learn more about gardening) |
Generativity | To make a lasting contribution to society |
Spirituality | To deepen one’s relationship with God or other religious entities |
Avoidance | To avoid letting anything upset you or get in the way |
Miscellaneous | Goals that do not belong to the above categories |
Results.
For each of the goal categories, we computed the proportion of goals in each category by summing the number of goals listed within each category and dividing by the total number of goals listed. We report the means and CIs across the seven goal categories by gender in Table S2.
Table S2.
Means with 95% CIs for the coded variables in study 1
Goal category | Female participants | Male participants |
Power | 0.03 [0.02, 0.04] | 0.07 [0.06, 0.08] |
Achievement | 0.31 [0.28, 0.33] | 0.30 [0.28, 0.33] |
Affiliation | 0.15 [0.13, 0.17] | 0.14 [0.13, 0.16] |
Personal growth and health | 0.32 [0.30, 0.35] | 0.32 [0.29, 0.34] |
Generativity | 0.04 [0.03, 0.05] | 0.04 [0.03, 0.05] |
Spirituality | 0.03 [0.02, 0.04] | 0.03 [0.02, 0.04] |
Avoidance | 0.01 [0.01, 0.02] | 0.01 [0.01, 0.01] |
Miscellaneous | 0.12 [0.10, 0.13] | 0.09 [0.07, 0.10] |
Owing to the significant difference in total goals listed by male and female participants, these data are presented as proportions.
Women listed more goals than men did [meanf = 9.46, SDf = 5.63 vs. meanm = 8.41, SDm = 5.28, t(779) = 2.67, P = 0.008, d = 0.19]. Compared to male participants, female participants spontaneously listed a smaller proportion of power-related goals out of their list of total goals [meanf = 3%, SDf = 9% vs. meanm = 7%, SDm = 14%, t(779) = 4.18, P < 0.001, d = 0.34]. There was also a main effect of gender on miscellaneous goals, such that women reported a higher proportion of such goals than did men [meanf = 12%, SDf = 17% vs. meanm = 9%, SDm = 16%, t(779) = 2.60, P = 0.01, d = 0.18]. We found no gender differences along the other goal dimensions: affiliation, achievement, personal growth and health, generativity, spirituality, or avoidance.
To further examine the distributions of answers for the number of goals men and women listed, we conducted nonparametric tests and analyzed various properties of the distributions (Table S3). The nonparametric tests confirmed that the number of overall goals women listed was statistically significantly higher than that of men (Mann–Whitney test, Z = −2.78, P = 0.005; Kolmogorov–Smirnov, Z = 1.38, P = 0.043).
Table S3.
Property of the distributions of answers in total goals listed by male and female participants in study 1
Property | Female participants | Male participants |
N | 330 | 451 |
Minimum–maximum | 1–25 | 1–25 |
Mean (SE) | 9.46 (0.31) | 8.41 (0.25) |
SD | 5.63 | 5.28 |
Variance | 31.67 | 27.91 |
Skewness (SE) | 1.07 (0.13) | 1.41 (0.12) |
Kurtosis (SE) | 0.61 (0.27) | 1.93 (0.23) |
Study 2
Method.
Participants first read initial instructions welcoming them to the study and answered an attention check. Those who failed the attention check were automatically taken to a screen letting them know that, based on their answers, they did not qualify for the study. Thus, their data were not recorded.
Next, participants were asked to list their core goals (as in study 1). We gave them 20 open text boxes and asked them to write each goal in a box. They could list anywhere from 1 to 20 goals; we did not ask them to rank the goals in order of importance but simply to list them in the order in which they came to mind. They were given 2 min to list their goals.
Participants were then presented with the goals they had listed and asked to indicate the category to which the goals belonged, using the same set of categories as in study 1. After this task, participants were asked to list favorite foods under the same 2-min time limit.
Finally, after answering demographic questions, participants indicated the extent to which, in completing the study, they were trying to please the experimenter (i.e., “the person who prepared the survey and who was collecting the data”) on a 7-point scale (1 = not at all, 7 = very much so).
Results.
We report the means and CIs across the seven goal categories by gender in Table S4. Compared to men, women listed a smaller proportion of power-related goals out of their list of total goals [meanf = 5%, SDf = 11% vs. meanm = 10%, SDm = 13%, t(435) = 4.51, P < 0.001, d = 0.42]. There was also a main effect of gender on avoidance goals such that women reported a higher proportion of such goals than did men [meanf = 2%, SDf = 7% vs. meanm = 1%, SDm = 4%, t(435) = 2.15, P = 0.032, d = 0.18]. We found no gender differences along the other goal dimensions.
Table S4.
Means with 95% CIs for the coded variables in study 2
Goal category | Female participants | Male participants |
Power | 0.05 [0.03, 0.06] | 0.10 [0.08, 0.12] |
Achievement | 0.27 [0.24, 0.30] | 0.28 [0.25, 0.31] |
Affiliation | 0.17 [0.14, 0.19] | 0.16 [0.14, 0.19] |
Personal growth and health | 0.27 [0.24, 0.30] | 0.24 [0.21, 0.27] |
Generativity | 0.04 [0.03, 0.05] | 0.05 [0.04, 0.06] |
Spirituality | 0.05 [0.04, 0.07] | 0.06 [0.04, 0.07] |
Avoidance | 0.02 [0.01, 0.03] | 0.01 [0.005, 0.02] |
Miscellaneous | 0.12 [0.10, 0.15] | 0.10 [0.08, 0.13] |
Owing to the significant difference in total goals listed by male and female participants, these data are presented as proportions.
Female participants again listed more goals than did male participants [meanf = 9.47, SDf = 4.81 vs. meanm = 7.90, SDm = 4.63, t(435) = 3.48, P = 0.001, d = 0.34]. Importantly, female and male participants listed about the same number of favorite foods [meanf = 13.54, SDf = 5.21 vs. meanm = 12.76, SDm = 5.70, t(435) = 1.50, P = 0.14, d = 0.14] and were similarly disinterested in pleasing the experimenter during the study [meanf = 2.69, SDf = 2.06 vs. meanm = 2.68, SDm = 1.91, t(435) < 1, P = 0.96, d = 0.005]. Even when controlling for the number of favorite foods respondents reported, women still listed more life goals than men, F(1,434) = 10.23, P = 0.001, = 0.023.
As in study 1, we further examine the distributions of answers for the number of goals men and women listed by conducting nonparametric tests and analyzing various properties of the distributions (Table S5). The nonparametric tests confirmed that the number of overall goals women listed was statistically significantly higher than that of men (Mann–Whitney test, Z = −3.96, P < 0.001; Kolmogorov–Smirnov, Z = 1.80, P = 0.003).
Table S5.
Property of the distributions of answers in total goals listed by male and female participants in study 2
Property | Female participants | Male participants |
N | 220 | 217 |
Minimum–maximum | 1–20 | 1–20 |
Mean (SE) | 9.47 (0.32) | 7.90 (0.31) |
SD | 4.81 | 4.63 |
Variance | 23.12 | 21.43 |
Skewness (SE) | 0.81 (0.16) | 1.19 (0.17) |
Kurtosis (SE) | −0.027 (0.33) | 0.94 (0.33) |
Study 3
Method.
After reading initial instructions welcoming them to the online study, participants were shown a ladder with rungs numbered from 1 to 10 and provided with the following instructions:
Imagine that the ladder below reflects the hierarchy of professional advancement in the industry in which you currently work. At the top of the ladder are positions like CEO and CFO or Members of the Board (if applicable). At the bottom of the ladder are jobs that are usually acquired immediately after undergrad, such as Analyst (for consulting or investment banking).
Now think about your own career. Please tell us where you think you are on the ladder at the moment, given your current job.
Participants indicated a ladder rung between 1 and 10 and then were prompted to provide the title of their current position in an open text box. On the next page, participants were asked to use the same ladder to indicate their ideal position and report the title of this position in an open text box. The specific instructions for this part were as follows:
What is the highest position that you would want to reach in your career, but not go beyond? In other words, what is your ideal position?
In the third part of the study, participants were asked to use the same ladder to indicate the “highest position that is realistically attainable for you (regardless of what your ideal position is)” and write the title of this position. Finally, participants answered demographic questions about their age, gender, and ethnicity.
Results.
There were no significant differences between men and women in the current position that they reported [meanf = 5.39, SDf = 1.85 vs. meanm = 5.63, SDm = 2.11, t(633) = −1.46, P = 0.145 d = 0.12]. To examine gender differences in the level of ideal position and in the level of attainable position, we conducted one-way ANOVAs controlling for respondents’ current position. We found that compared to male participants, female participants reported a lower ideal position [meanf = 9.04, SDf = 1.14 vs. meanm = 9.59, SDm = 0.92, F(1,632) = 41.99, P < 0.001, = 0.062]. However, they reported equally high levels of their highest attainable position [meanf = 9.29, SDf = 0.88 vs. meanm = 9.41, SDm = 0.98, F(1,632) = 1.58, P = 0.21, = 0.002).
Study 4
Method.
We asked participants to “think about the things you’d like to accomplish in your life and the goals you have for yourself.” We asked participants to think about two power-related goals: (i) “As one of my core goals in life, I would like to have a powerful position in an organization” and (ii) “As one of my core goals in life, I would like to have power over others.” We then asked them to rate the desirability (α = 0.82) and attainability (α = 0.85) of these two goals on 7-point Likert scales.
Results.
Compared to male participants, female participants rated these goals as less desirable (meanf = 3.73, SDf = 1.67, 95% CI = [3.42, 4.05] vs. meanm = 4.19, SDm = 1.52, 95% CI = [3.93, 4.45], t(245) = 2.24, P = 0.026, d = 0.29). However, women and men believed these goals to be equally attainable (meanf = 4.25, SDf = 1.67, 95% CI = [3.94, 4.57] vs. meanm = 4.35, SDm = 1.56, 95% CI = [4.09, 4.62], t(245) = 0.49, P = 0.63, d = 0.06).
Study 5
Method.
Participants read initial instructions welcoming them to the study and answered two attention checks. Those who failed either attention check were automatically taken to a screen letting them know that, based on their answers, they did not qualify for the study. Thus, their data were not recorded. We then asked participants to read a hypothetical scenario and answer a few questions about it. The scenario read as follows:
You are satisfied with your current job, but there is the possibility for you to be promoted to a higher-level position in the organization where you work. As a result of this promotion, your level of power over others would increase substantially.
We asked participants to indicate how much they thought they would experience different outcomes if they received the promotion, using a 10-point scale (1 = not at all to 10 = a lot). The outcomes were (i) status or influence, (ii) stress or anxiety, (iii) time constraints, (iv) burden of responsibility, (v) satisfaction or happiness, (vi) money, (vii) difficult tradeoffs or sacrifice, (viii) opportunity, and (ix) conflict with other life goals.
Participants were then asked to indicate how desirable they found such a promotion to be on a 7-point scale (1 = not at all to 7 = very much). Next, we asked them “If the chance to get such type of promotion was available to you right now, but required effort on your part, would you go after the promotion?” They answered this question using a 7-point scale (1 = definitely not to 7 = definitely yes). Finally, participants answered demographic questions (age and gender).
Results.
Data exclusions.
Ten participants took the survey more than once. We excluded them from the analyses according to a priori exclusion rules we made before conducting the study.
Main analyses.
We conducted exploratory factor analyses to find out how the items assessing participants’ expectations regarding the outcomes associated with the potential promotion loaded together. The items loaded onto two factors (Table S6), one representing items connected with positive outcomes (α = 0.77) and the other representing items connected with negative outcomes (α = 0.84).
Table S6.
Rotated component matrix of positive and negative outcomes in study 5
Outcome associated with promotion | Factor 1 | Factor 2 |
Stress or anxiety | 0.813 | −0.020 |
Difficult tradeoffs or sacrifice | 0.809 | 0.008 |
Time constraints | 0.782 | 0.154 |
Burden of responsibility | 0.781 | 0.194 |
Conflict with other life goals | 0.695 | −0.064 |
Satisfaction or happiness | 0.071 | 0.803 |
Opportunity | 0.119 | 0.798 |
Money | −0.221 | 0.763 |
Status or influence | 0.181 | 0.698 |
Extraction method: principal component analysis. Rotation method: varimax with Kaiser normalization.
We report the means and confidence intervals for the means for the variables measured in the study by gender in Table S7. Female participants reported anticipating more negative outcomes to the potential promotion (meanf = 6.73, SD = 1.56) than did male participants (meanm = 6.23, SD = 1.39), t(442) = 3.43, P = 0.001, d = 0.34. However, they reported about the same amount of positive outcomes (meanf = 7.13, SD = 1.44 vs. meanm = 7.23, SD = 1.28), t(442) = 0.79, P = 0.43, d = 0.07.*
Table S7.
Means with 95% confidence intervals for the variables assessed in study 5 by gender
Variable | Female participants | Male participants |
Desirability of the promotion | 5.12 [4.89, 5.36] | 5.48 [5.33, 5.63] |
Likelihood of pursuing the promotion | 4.77 [4.55, 5.00] | 5.25 [5.08, 5.41] |
Negative outcomes associated with the promotion | 6.73 [6.48, 6.98] | 6.23 [6.07, 6.40] |
Positive outcomes associated with the promotion | 7.13 [6.90, 7.35] | 7.23 [7.08, 7.38] |
We also computed a dichotomous measure for whether participants anticipated more negative than positive outcomes to be associated with the potential promotion (1 if they did, 0 otherwise). More female than male participants [37.7% (58/154) vs. 24.5% (71/290)] anticipated greater negative outcomes to the potential promotion, χ2(1,N = 444) = 8.48, P = 0.004, Cramér’s V = 0.14.
Female participants also reported viewing the potential promotion as less desirable (meanf = 5.12, SD = 1.46) than did male participants (meanm = 5.48, SD = 1.32), t(442) = 2.60, P = 0.01, d = 0.26, and, compared to male participants, indicated that they would be less likely to pursue the promotion (meanf = 4.77, SD = 1.41 vs. meanm = 5.25, SD = 1.43), t(442) = 3.35, P = 0.001, d = 0.34.
Mediation analyses.
We predicted that the negative outcomes women believed they would experience after achieving a position of high power explains their lower likelihood to pursue the promotion. In a regression using likelihood of promotion pursuit as the dependent measure, negative outcomes associated with the potential promotion as the mediator, and positive outcomes associated with the same promotion as the control variable, the effect of gender weakened (from β = 0.14, P = 0.001 to β = 0.098, P = 0.011), and negative outcomes associated with the potential promotion predicted a lower likelihood of going after the promotion (β = −0.24, P < 0.001). Bootstrap analyses with 10,000 iterations (3) showed that the 95% bias-corrected confidence interval for the size of the indirect effect for negative outcomes associated with the potential promotion excluded zero (0.05, 0.21), indicating a significant indirect effect.
We conducted similar analyses using desirability of the promotion as the dependent measure. When including expected negative outcomes to the potential promotion as the mediator, the effect of gender weakened (from β = 0.10, P = 0.008 to β = 0.046, P = 0.19), and negative outcomes associated with the potential promotion predicted a lower desirability of the promotion (β = −0.33, P < 0.001; 95% bias-corrected CI = [0.07, 0.26]).
Mediation analyses only considering items related to difficult tradeoffs and conflict with other life goals.
We repeated the same analyses using only two of the expected negative outcomes: difficult tradeoffs and conflict with other life goals (α = 0.70). In a regression using likelihood of promotion pursuit as the dependent measure, these two items assessing conflict as the mediator, and positive outcomes associated with the same promotion as the control variable, the effect of gender weakened (from β = 0.14, P = 0.001 to β = 0.11, P = 0.004), and our measure of conflict predicted a lower likelihood of going after the promotion (β = −0.21, P < 0.001). The 95% bias-corrected confidence interval for the size of the indirect effect for conflict excluded zero (0.02, 0.16), indicating a significant indirect effect.
We conducted similar analyses using desirability of the promotion as the dependent measure. When including the two-item measure of conflict as the mediator, the effect of gender weakened (from β = 0.10, P = 0.008 to β = 0.06, P = 0.083), and conflict predicted a lower desirability of the promotion (β = −0.33, P < 0.001; 95% bias-corrected CI = [0.02, 0.22]).
The nature and significance of these results do not change when considering only the single item “conflict with other life goals.”
Study 6
Method.
We asked participants to answer a short questionnaire, which included the same scenario and measures as in study 5. They also answered demographic questions about their age and gender.
Results.
We report the means and confidence intervals for the variables measured in this study in Table S8. As in study 5, we aggregated the outcomes executives associated with the potential promotion into two factors: positive (α = 0.73) and negative (α = 0.65) outcomes. Female executives expected more negative outcomes to be associated with the promotion (meanf = 7.09, SD = 1.71) than did male executives (meanm = 6.17, SD = 2.33), t(202) = 2.61, P = 0.01, d = 0.45. However, they anticipated similar levels of positive outcomes (meanf = 7.39, SD = 1.38 vs. meanm = 7.11, SD = 1.43), t(202) = 1.22, P = 0.22, d = 0.20.†
Table S8.
Means with 95% confidence intervals for the variables assessed in study 6 by gender
Variable | Female participants | Male participants |
Desirability of the promotion | 5.69 [5.37, 6.01] | 6.02 [5.87, 6.17] |
Likelihood of pursuing the promotion | 5.71 [5.34, 6.08] | 6.07 [5.88, 6.25] |
Negative outcomes associated with the promotion | 7.09 [6.61, 7.56] | 6.17 [5.79, 6.54] |
Positive outcomes associated with the promotion | 7.39 [7.01, 7.77] | 7.11 [6.88, 7.34] |
We also computed a dichotomous measure for whether participants expected more negative than positive outcomes with the potential promotion (1 if they did, 0 otherwise). Female participants expected more negative outcomes than did male participants [46.2% (24/52) vs. 25.7% (39/152), respectively] to the potential promotion, χ2(1,N = 204) =7.63, P = 0.006, Cramér’s V = 0.19.
Female participants also reported viewing the potential promotion as less desirable (meanf = 5.69, SD = 1.15) than did male participants, (meanm = 6.02, SD = 0.91), t(202) = 2.09, P = 0.038, d = 0.32, and, compared to the male participants, indicated that they would be less likely to pursue the promotion (meanf = 5.71, SD = 1.33 vs. meanm = 6.07, SD = 1.15), t(202) = 1.84, P = 0.067, d = 0.29.
Mediation analyses.
We conducted regression analyses with likelihood of promotion pursuit as the dependent measure, negative outcomes associated with the potential promotion as the potential mediator, and positive outcomes associated with the promotion as the control variable. We found that the effect of gender weakened (from β = 0.15, P = 0.029 to β = 0.109, P = 0.104), and negative outcomes associated with the potential promotion predicted a lower likelihood of going after the promotion (β = −0.27, P < 0.001). Bootstrap analyses with 10,000 iterations (3) showed that the 95% bias-corrected confidence interval for the size of the indirect effect excluded zero (0.01, 0.31), indicating a significant indirect effect.
We conducted similar analyses using desirability of the promotion as the dependent measure. When negative outcomes associated with the potential promotion was included as the mediator, the effect of gender weakened (from β = 0.17, P = 0.01 to β = 0.13, P = 0.044), and negative outcomes associated with the potential promotion predicted a lower desirability of the promotion (β = −0.28, P < 0.001; 95% bias-corrected CI = [0.01, 0.26]).
Mediation analyses only considering items related to difficult tradeoffs and conflict with other life goals.
We repeated the same analyses using only two of the expected negative outcomes: difficult tradeoffs and conflict with other life goals (α = 0.74). In a regression using likelihood of promotion pursuit as the dependent measure, these two items assessing conflict as the mediator, and positive outcomes associated with the same promotion as the control variable, the effect of gender weakened (from β = 0.15, P = 0.029 to β = 0.08, P = 0.23), and our measure of conflict predicted a lower likelihood of going after the promotion (β = −0.38, P < 0.001). The 95% bias-corrected confidence interval for the size of the indirect effect for conflict excluded zero (0.07, 0.35), indicating a significant indirect effect.
We conducted similar analyses using desirability of the promotion as the dependent measure. When including the two-item measure of conflict as the mediator, the effect of gender weakened (from β = 0.17, P = 0.01 to β = 0.09, P = 0.13), and conflict predicted a lower desirability of the promotion (β = −0.42, P < 0.001; 95% bias-corrected CI = [0.06, 0.31]).
The nature and significance of these results do not change when considering only the single item “conflict with other life goals.”
Study 7
Method.
We asked participants to answer a short questionnaire, which included an adapted version of the scenario presented to participants in studies 5 and 6. Instead of being asked to imagine being promoted to a position above their current one, participants in this study were asked to imagine that, upon graduation, they were presented with a high-power job opportunity. The instructions read as follows:
Soon after graduation, you have the opportunity to take a position in an organization that will give you a lot of power over other people and resources. In thinking about this position, what are the primary outcomes or feelings you associate with it (e.g., opportunity, difficult tradeoffs)?
We asked participants to describe the high-power job that they were imagining and then list the primary outcomes or feelings they associated with taking such a position. They were provided with 15 open text boxes in which to write each outcome; they were allowed to list anywhere from 1 to 15 items.
We also asked participants to indicate on 7-point scales (i) how desirable the position would be to them, (ii) their likelihood of pursuing the position if it required extra effort, and (iii) their likelihood of pursuing the position if it did not require any extra effort.
Next, we asked participants to look back to the outcomes they anticipated being associated with to the high-power job they had been asked to imagine. They had to indicate the nature of these anticipated outcomes, using the categories we provided (i.e., positive, negative, or neutral).
Finally, participants answered demographic questions about their age and gender.
Fourteen participants were excluded from the analyses for failing to appropriately complete the study. Examples of such noncompliance include writing “good” as the job being imagined and listing “none” as the outcomes associated with the job. They also did not provide information regarding their gender.
Results.
For each participant, we computed the proportion of outcomes expected with the promotion of each type (positive, neutral, or negative) by summing the number of outcomes listed within each type and dividing by the total number of outcomes listed.
We report the means and confidence intervals by gender in Table S9. Compared to men, women listed a lower total number of outcomes they associated with the high-power position [meanf = 3.84, SDf = 2.61 vs. meanm = 4.93, SDm = 2.75, t(514) = 4.60, P < 0.001, d = 0.41]. Compared to male participants, female participants spontaneously listed a lower proportion of positive outcomes [meanf = 50.1%, SDf = 38.9% vs. meanm = 56.6%, SDm = 32.9%, t(514) = 2.05, P = 0.041, d = 0.18] and a higher proportion of negative outcomes [meanf = 30%, SDf = 37% vs. meanm = 22.5%, SDm = 25.8%, t(514) = 2.65, P = 0.008, d = 0.24]. The proportion of neutral outcomes was the same for men and women [meanf = 19.9%, SDf = 29.9% vs. meanm = 20.9%, SDm = 24.2%, t(514) = 0.39, P = 0.70, d = 0.03].
Table S9.
Means with 95% confidence intervals for the variables assessed in study 7 by gender
Variable | Female participants | Male participants |
Desirability of the promotion | 5.02 [4.89, 5.15] | 5.37 [5.19, 5.55] |
Likelihood of going after the promotion when extra effort is required | 5.07 [4.93, 5.20] | 5.41 [5.23, 5.59] |
Likelihood of going after the promotion when no extra effort is required | 5.54 [5.41, 5.67] | 5.97 [5.82, 6.11] |
Proportion of negative outcomes associated with the promotion | 0.30 [0.26, 0.34] | 0.22 [0.19, 0.25] |
Proportion of positive outcomes associated with the promotion | 0.50 [0.45, 0.55] | 0.57 [0.52, 0.61] |
Female participants also reported viewing the high-power position as less desirable than male participants [meanf = 5.02, SDf = 1.09 vs. meanm = 5.37, SDm = 1.43, t(514) = 3.16, P = 0.002, d = 0.28]. Compared to men, women were less likely to pursue the position, regardless of whether it necessitated extra effort on their part [meanf = 5.07, SDf = 1.14 vs. meanm = 5.41, SDm = 1.41, t(514) = 3.09, P = 0.002, d = 0.27] or not [meanf = 5.54, SDf = 1.09 vs. meanm = 5.97, SDm = 1.15, t(514) = 4.29, P < 0.001, d = 0.38].
We conducted regression analyses with likelihood of promotion pursuit as the dependent measure, proportion of negative outcomes associated with the potential promotion as the potential mediator, and gender as our independent variable. For our dependent measure, we averaged participants’ answers to the two items assessing their interest in pursuing the promotion (α = 0.80). We found that the effect of gender weakened (from β = 0.17, P < 0.001 to β = 0.14, P = 0.001), and the proportion of negative outcomes associated with the potential promotion predicted a lower likelihood of going after the promotion (β = −0.28, P < 0.001). Bootstrap analyses with 10,000 iterations (3) showed that the 95% bias-corrected confidence interval for the size of the indirect effect excluded zero (0.02, 0.13), indicating a significant indirect effect.
We conducted similar analyses using desirability of the promotion as the dependent measure. When the proportion of negative outcomes associated with the potential promotion was included as the mediator, the effect of gender weakened (from β = 0.14, P = 0.02 to β = 0.10, P = 0.018), and the proportion of negative outcomes associated with the potential promotion predicted a lower desirability of the promotion (β = −0.35, P < 0.001; 95% bias-corrected CI = [0.03, 0.18]).
Study 8
Method.
Participants were randomly assigned to one of two conditions: (i) promotion with power or (ii) promotion with power defined. Participants first read initial instructions welcoming them to the study and answered two attention checks. Those who failed either attention check were automatically taken to a screen letting them know that, based on their answers, they did not qualify for the study. Thus, their data were not recorded. Participants were then asked to read a hypothetical scenario and answer a few questions about it. In the promotion with power condition, the scenario read as follows:
You are satisfied with your current job, but there is the possibility for you to be promoted to a higher-level position in the organization where you work. As a result of this promotion, your level of power would increase substantially.
In the promotion with power defined condition, the scenario read as follows:
You are satisfied with your current job, but there is the possibility for you to be promoted to a higher-level position in the organization where you work. As a result of this promotion, your level of power would increase substantially. By power, we mean your relative ability to control other people’s outcomes, experiences, or behaviors.
Then, in both conditions, participants were told, “Please imagine you in fact received this promotion.” We then asked participants to indicate the extent to which they thought such a promotion would create conflicts with their other life goals (1 = not at all, 7 = very much) and the extent to which they thought such a promotion would require them to make tradeoffs and sacrifices given their other life goals (1 = not at all, 7 = very much). We used these items as a measure of expected conflict (α = 0.85). Participants were then asked to indicate how desirable they would find such a promotion to be on a 7-point scale (1 = not at all, to 7 = very much).
Next, participants were asked, “Now please think about having power in a job more generally. Which words do you associate with such a concept? (Please list 2 or 3 words below).” They were also asked to answer the question “How do you view having power in a job?” They could choose between two possible answers: (i) “A goal that I am not that interested in pursuing” or (ii) “A goal I definitely want to pursue.” Finally, participants answered a few demographic questions.
Results.
Table S10 reports the means and confidence intervals for the means for the variables measured in the study by gender.
Table S10.
Means with 95% confidence intervals for the variables assessed in study 8 by gender
Variable | Male participants | Female participants | |
Desirability of the promotion | 5.45 [5.30, 5.60] | 5.18 [4.96, 5.40] | |
Predicted conflict the promotion would bring about | 3.41 [3.25, 3.56] | 3.80 [3.58, 4.04] |
Outcomes associated with the potential promotion.
We conducted a 2 (male vs. female participant) × 2 (power defined vs. power not defined) ANOVA using expected conflict as the dependent measure. This analysis revealed an insignificant effect for the power-defined manipulation [F(1,480) = 1.44, P = 0.23, = 0.003] and an insignificant interaction [F(1,480) = 1.15, P = 0.28, = 0.002]. However, gender significantly predicted expected conflict, F(1,480) = 8.26, P = 0.004, = 0.02: Female participants reported the promotion would make them experience greater conflict (mean = 3.81, SD = 1.58) compared to male participants (mean = 3.41, SD = 1.37).
We conducted a similar 2 × 2 ANOVA using desirability of the promotion as the dependent measure. This analysis revealed a marginally significant effect for the power-defined manipulation [F(1,480) = 3.13, P = 0.08, = 0.005] and an insignificant interaction [F(1,480) < 1, P = 0.98, = 0.000]. However, gender significantly predicted the desirability of the promotion, F(1,480) = 4.36, P = 0.037, = 0.01: Female participants reported viewing the promotion as less desirable (mean = 5.18, SD = 1.50) than did male participants (mean = 5.45, SD = 1.32).
Mediation analyses.
We predicted that the predicted conflict women believe they would experience when achieving a position of high power explains their lower ratings for how desirable such a position is. In a regression using desirability of the promotion as the dependent measure, predicted conflict as potential mediator, and the power-defined manipulation as control, the effect of gender weakened (from β = 0.095, P = 0.037 to β = 0.031, P = 0.44), and conflict predicted lower ratings for how desirable the promotion would be (β = −0.49, P < 0.001). Bootstrap analyses with 10,000 iterations (3) showed that the 95% bias-corrected confidence interval for the size of the indirect effect for predicted conflict excluded zero (0.06, 0.32), indicating a significant indirect effect.
Power as a goal.
We also asked participants to indicate whether power was a goal they wanted to pursue or not. More female than male participants [41.0% (75/183) vs. 30.9% (93/301)] indicated not being interested in pursuing power as a goal, χ2(1,N = 484) = 5.11, P = 0.024, Cramér’s V = 0.10.
Words associated with power.
Four gender-blind coders, two female and two male, coded the words participants associated with having power at work. Each word was assigned a code of positive, negative, or neutral, and we averaged across the codes provided by all four coders for our analyses (all α > 0.70, average α = 0.83). There were no gender differences in the number of positive words [meanf = 1.49, SDf = 0.90 vs. meanm = 1.52, SDm = 0.79, F(1,483) = 0.24, P = 0.627, = 0.000] or neutral words [meanf = 0.79, SDf = 0.57 vs. meanm = 0.83, SDm = 0.57, F(1,483) = 0.68, P = 0.41, = 0.001] listed by participants, but female participants listed significantly more negative words associated with having power at work [meanf = 0.50, SDf = 0.73 vs. meanm = 0.37, SDm = 0.65, F(1,483) = 4.56, P = 0.033, = 0.009].
Study 9
Method.
Participants were assigned to one of three promotion description conditions: (i) basic promotion, (ii) promotion with power, or (iii) promotion with power defined. The prompt in the basic promotion condition read as follows:
Please imagine the following scenario. There is a possibility for you to be promoted to the position just above your current job in the organization where you work.
The prompt in the promotion with power condition read as follows:
Please imagine the following scenario. There is a possibility for you to be promoted to the position just above your current job in the organization where you work. As a result of this promotion, your level of power would increase substantially.
Finally, the prompt in the promotion with power defined condition read as follows:
Please imagine the following scenario. There is a possibility for you to be promoted to the position just above your current job in the organization where you work. As a result of this promotion, your level of power would increase substantially. By power, we mean your ability to control other people’s outcomes, experiences, or behaviors.
Participants were then presented with the same nine items as in studies 5–7 and asked to answer “How much do you think you would experience each of the outcomes below if you received this promotion?” on a scale from 1 to 7 (1 = not at all, 7 = a lot). Participants then used 7-point scales to report (i) “How desirable would you find a promotion at your current job to be?” (1 = not at all, 7 = very much) and (ii) “If the chance to get a promotion was available to you right now, but required a lot of effort on your part, would you go after the promotion?” (1 = definitely not, 7 = definitely yes). Finally, participants reported basic demographics.
Results.
We conducted 2 (men vs. women) × 3 (basic promotion vs. promotion with power vs. promotion with power defined) ANOVAs on our main variables of interest and found no main effects for the promotion description manipulation or significant interactions (all P > 0.11). Across our analyses, the only significant effect was that for gender.
Compared to male participants, female participants expected more negative outcomes with the promotion [meanf = 5.68, SDf = 1.13 vs. meanm = 5.02, SDm = 1.46; F(1,259) = 15.89, P < 0.001, = 0.06]. However, men and women expected a statistically equivalent level of positive outcomes with the promotion [meanf = 6.15, SDf = 0.88 vs. meanm = 6.26, SDm = 0.82; F(1,259) = 1.36, P = 0.25, = 0.005]. Female participants also reported viewing the potential promotion as less desirable than did men [meanf = 5.35, SDf = 1.38 vs. meanm = 6.05, SDm = 1.12; F(1,259) = 20.90, P < 0.00, = 0.075] and indicated that they would be less likely to go after it [meanf = 4.98, SDf = 1.51 vs. meanm = 5.83, SDm = 1.36; F(1,259) = 22.93, P < 0.001, = 0.81].
Acknowledgments
The authors thank Hannah Riley Bowles, Robin Ely, Brian Hall, Erin Hennes, Kathleen McGinn, Todd Rogers, Elizabeth Baily Wolf, and Ting Zhang for insightful feedback on earlier versions of this manuscript, as well as Alind Amedi, Elizabeth Grether, Ethan Ludwin-Peery, Surya Menon, Helena Rabasco, and Mindi Rock for valuable research assistance.
Footnotes
The authors declare no conflict of interest.
This article is a PNAS Direct Submission. S.J.C. is a guest editor invited by the Editorial Board.
*We note that when considering each of the items assessing negative and positive outcomes associated with the potential promotion separately, we find significant gender differences on all the negative items except for time constraints (where P = .12), and only on one positive item [status or influence, P = .023 (such that men expect greater status and influence than women do); all other P > .38].
†We note that when considering each of the items assessing negative and positive outcomes associated with the potential promotion separately, we find significant gender differences on all the negative items except for time constraints (where P = .24) and responsibility (where P = .38) and on none of the positive items at the 5% level (satisfaction or happiness: P = .085 and opportunity: P = .10; both were higher for women than for men).
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1502567112/-/DCSupplemental.
References
- 1.Burke RJ, Major DA. Gender in Organizations: Are Men Allies or Adversaries to Women’s Career Advancement? Edward Elgar; Cheltenham, UK: 2014. [Google Scholar]
- 2. Catalyst (2014). Catalyst charts growth of women on America’s corporate boards (Catalyst, New York)
- 3.National Science Foundation 2013. Women, minorities, and persons with disabilities in science and engineering (National Science Foundation, Arlington, VA)
- 4.Brush CG, Greene PG, Balachandra L, Davis AE. 2014. The Diana Report. Women entrepreneurs 2014: Bridging the gender gap in venture capital (Arthur M. Blank Center for Entrepreneurship, Babson College, Wellesley, MA)
- 5. Credit Suisse Research Institute (2014). The CS gender 3000: Women in senior management. Available at https://publications.credit-suisse.com/tasks/render/file/index.cfm?fileid=8128F3C0-99BC-22E6-838E2A5B1E4366DF. Accessed July 22, 2015.
- 6.Williams WM, Ceci SJ. National hiring experiments reveal 2:1 faculty preference for women on STEM tenure track. Proc Natl Acad Sci USA. 2015;112(17):5360–5365. doi: 10.1073/pnas.1418878112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Heilman ME, Eagly AH. Gender stereotypes are alive, well, and busy producing workplace discrimination. Ind Organ Psychol. 2008;1:393–398. [Google Scholar]
- 8.Heilman ME, Parks-Stamm EJ. Gender stereotypes in the workplace: Obstacles to women’s career progress. In: Correll SJ, editor. Social Psychology of Gender: Advances in Group Processes. Vol 24. JAI; Oxford: 2007. pp. 47–78. [Google Scholar]
- 9.Moss-Racusin CA, Dovidio JF, Brescoll VL, Graham MJ, Handelsman J. Science faculty’s subtle gender biases favor male students. Proc Natl Acad Sci USA. 2012;109(41):16474–16479. doi: 10.1073/pnas.1211286109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brescoll V, Uhlmann EL. Can an angry woman get ahead? Status conferral, gender, and expression of emotion in the workplace. Psychol Sci. 2008;19(3):268–275. doi: 10.1111/j.1467-9280.2008.02079.x. [DOI] [PubMed] [Google Scholar]
- 11.Brooks AW, Huang L, Kearney SW, Murray FE. Investors prefer entrepreneurial ventures pitched by attractive men. Proc Natl Acad Sci USA. 2014;111(12):4427–4431. doi: 10.1073/pnas.1321202111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Butler D, Geis FL. Nonverbal affect responses to male and female leaders: Implications for leadership evaluations. J Pers Soc Psychol. 1990;58:48–59. [Google Scholar]
- 13.Heilman ME, Block CJ, Martell RF, Simon MC. Has anything changed? Current characterization of men, women, and managers. J Appl Psychol. 1989;74:935–942. [Google Scholar]
- 14.Eagly AH, Karau SJ. Role congruity theory of prejudice toward female leaders. Psychol Rev. 2002;109(3):573–598. doi: 10.1037/0033-295x.109.3.573. [DOI] [PubMed] [Google Scholar]
- 15.Eagly AH, Karau SJ, Makhijani MG. Gender and the effectiveness of leaders: A meta-analysis. Psychol Bull. 1995;117(1):125–145. doi: 10.1037/0033-2909.117.1.125. [DOI] [PubMed] [Google Scholar]
- 16.Reuben E, Sapienza P, Zingales L. How stereotypes impair women’s careers in science. Proc Natl Acad Sci USA. 2014;111(12):4403–4408. doi: 10.1073/pnas.1314788111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Eagly AH, Steffen VJ. 1986. Gender and aggressive behavior: A meta-analytic review of the social psychological literature. Sex Differences in Social Behavior: A Social Role Interpretation, Eagly AH (Lawrence Erlbaum, Hillsdale, NJ), pp 128–135. [PubMed]
- 18.Archer J. Does sexual selection explain human sex differences in aggression? Behav Brain Sci. 2009;32(3-4):249–266, discussion 266–311. doi: 10.1017/S0140525X09990951. [DOI] [PubMed] [Google Scholar]
- 19.Eagly AH, Steffen VJ. Gender and aggressive behavior: A meta-analytic review of the social psychological literature. Psychol Bull. 1986;100(3):309–330. [PubMed] [Google Scholar]
- 20.Dovidio JF, Ellyson SL, Keating CF, Heltman K, Brown CE. The relationship of social power to visual displays of dominance between men and women. J Pers Soc Psychol. 1988;54:233–242. doi: 10.1037//0022-3514.54.2.233. [DOI] [PubMed] [Google Scholar]
- 21.LaFrance M, Hecht MA, Levy Paluck E. The contingent smile: A meta-analysis of sex differences in smiling. Psychol Bull. 2003;129:305–334. doi: 10.1037/0033-2909.129.2.305. [DOI] [PubMed] [Google Scholar]
- 22.Swim JK. Perceived versus meta-analytic effect sizes: An assessment of the accuracy of gender stereotypes. J Pers Soc Psychol. 1994;66:21–36. [Google Scholar]
- 23.Bowles HR, Babcock L, Lai L. Social incentives for gender differences in the propensity to initiate negotiation: Sometimes it does hurt to ask. Organ Behav Hum Decis Process. 2007;103:84–103. [Google Scholar]
- 24.Eagly AH, Karau SJ. Gender and the emergence of leaders: A meta-analysis. J Pers Soc Psychol. 1991;60:685–710. [Google Scholar]
- 25.Buser T, Niderle M, Oosterbeek H. Gender, competitiveness, and career choices. Q J Econ. 2014;129(3):1409–1447. [Google Scholar]
- 26.Niederle M, Vesterlund L. Do women shy away from competition? Do men compete too much? Q J Econ. 2007;122(3):1067–1101. [Google Scholar]
- 27.Karoly P. A goal systems–self-regulatory perspective on personality, psychopathology, and change. Rev Gen Psychol. 1999;3:264–291. [Google Scholar]
- 28.Greene BA, DeBacker TK. Gender and orientation toward the future: Links to motivation. Educ Psychol Rev. 2004;16(2):91–120. [Google Scholar]
- 29.Mason MF, Zhang S, Dyer RL. Male susceptibility to attentional capture by power cues. J Exp Soc Psychol. 2010;46(2):482–485. [Google Scholar]
- 30.McClelland D. The Achieving Society. Free; New York: 1961. [Google Scholar]
- 31.Schuh SC, et al. Gender differences in leadership role occupancy: The mediating role of power motivation. J Bus Ethics. 2014;120:363–379. [Google Scholar]
- 32.Diener E, Fujita F. Resources, personal strivings, and subjective well-being: A nomothetic and idiographic approach. J Pers Soc Psychol. 1995;68(5):926–935. doi: 10.1037//0022-3514.68.5.926. [DOI] [PubMed] [Google Scholar]
- 33.Lubinski D, Benbow CP, Kell HJ. Life paths and accomplishments of mathematically precocious males and females four decades later. Psychol Sci. 2014;25(12):2217–2232. doi: 10.1177/0956797614551371. [DOI] [PubMed] [Google Scholar]
- 34.Ferriman K, Lubinski D, Benbow CP. Work preferences, life values, and personal views of top math/science graduate students and the profoundly gifted: Developmental changes and gender differences during emerging adulthood and parenthood. J Pers Soc Psychol. 2009;97(3):517–532. doi: 10.1037/a0016030. [DOI] [PubMed] [Google Scholar]
- 35.Hakim C. Women, careers, and work-life preferences. Br J Guidance Counseling. 2006;34:279–294. [Google Scholar]
- 36.Hakim C. Work-Lifestyle Choices in the 21st Century: Preference Theory. Oxford Univ Press; Oxford: 2000. [Google Scholar]
- 37.Emmons RA. Personal strivings: An approach to personality and subjective wellbeing. J Pers Soc Psychol. 1986;51:1058–1068. [Google Scholar]
- 38.Emmons RA. The Psychology of Ultimate Concerns: Motivation and Spirituality in Personality. Guilford; New York: 1999. [Google Scholar]
- 39.Emmons RA. Personal goals, life meaning, and virtue: Wellsprings of a positive life. In: Keyes C, Haidt J, editors. Flourishing: Positive Psychology and the Life Well-Lived. American Psychological Assoc; Washington, DC: 2003. pp. 105–128. [Google Scholar]
- 40.Brescoll VL. Who takes the floor and why: Gender, power, and volubility in organizations. Adm Sci Q. 2011;56:622–641. [Google Scholar]
- 41.Coffman KB. Evidence on self-stereotyping and the contribution of ideas. Q J Econ. 2014 doi: 10.1093/qje/qju023. [DOI] [Google Scholar]
- 42.Rudman LA, Glick P. Feminized management and backlash toward agentic women: The hidden costs to women of a kinder, gentler image of middle managers. J Pers Soc Psychol. 1999;77(5):1004–1010. doi: 10.1037//0022-3514.77.5.1004. [DOI] [PubMed] [Google Scholar]
- 43.Rudman LA, Moss-Racusin CA, Phelan JE, Nauts S. Status incongruity and backlash effects: Defending the gender hierarchy motivates prejudice against female leaders. J Exp Soc Psychol. 2012;48:165–179. [Google Scholar]
- 44.Hochschild A, Machung A. The Second Shift. Avon; New York: 1990. [Google Scholar]