Leveraging the Social Role of Dad to Change Gender Stereotypes of Men

Abstract

Trait stereotypes of men tend to be more fixed and negative than those of women. The current studies test whether stereotypes of men can be shifted through leveraging their social role as fathers. Trait attributes perceived to characterize women and moms were highly redundant, but those of men and dads were less so; moreover, men were perceived more negatively than dads, women, and moms (Study 1). Perceivers for whom the social role father was made salient rated men more similarly to dads, and no less similarly to men, and rated men more positively relative to a control condition (Study 2). Finally, among men, a threat to the category men resulted in greater opposition to benevolent social policies, but not if the social role father was primed (Study 3). Discussion focuses on positive consequences of increasing the psychological connection between men and fatherhood.

Keywords

social role theory stereotypes involved father manhood

According to social role theory (Eagly & Wood, 2012), observing members of a group as they take on particular social roles, and perform behaviors compelled by those roles, leads perceivers to view trait characteristics inferred from role-congruent behaviors as stereotypic of the group. Firefighters will be seen as brave, nurses as caring, and professors as intelligent, in part, because behaviors consistent with these trait characteristics are necessary for the successful performance of these roles. When members of social groups are numerically overrepresented in particular roles (women are nurses, men are firefighters), the trait characteristics implied by the behaviors required of that role come to be associated with the group itself. Thus, women come to be seen as particularly caring and nurturing, and men as brave and decisive.

In work examining stereotype change, Koenig and Eagly (2014) suggest that changes to social roles might be leveraged as a means for producing change in group stereotypes. To the extent that a group takes on a new or different role, and that role requires a different set of behaviors and their correspondent traits, this should produce change in the group stereotype. In Study 4, they presented participants with information about the future roles that specific groups were projected to take on by demographers over the next 25 to 30 years (e.g., “White men will be more common in the occupations of teachers, homemakers, and nurses”) and asked them to rate what these groups would be like in the future. They found that future group stereotypes shifted to incorporate the traits associated with these future roles (e.g., White men were predicted to be relatively high on communal traits in this particular condition).

While research on gender stereotypes tends to focus on women, a body of research suggests that stereotypes of men may be at least as—if not more—rigid as those of women, and problematic in their own way (see Croft, Schmader, & Block, 2015). Diekman and Eagly (2000) found that whereas women were perceived as having changed in their trait characteristics, becoming more masculine and less feminine, men were seen as relatively stable and high on masculinity, and low on femininity, and that pattern was expected to persist into the future. Twenge (1997) similarly reports in a meta-analysis that although women have increasingly endorsed masculine traits over time, men have not shown a parallel increase in their endorsement of feminine traits.

At the same time, researchers have identified problematic aspects of the male stereotype. Men are perceived in more negative terms than women (Eagly & Mladinic, 1989; Prentice & Carranza, 2002), and the content of the stereotype of men at times puts them at risk, for example, through ignoring medical or health problems, or difficulty in relationships due to a lack of emotional engagement (Moss-Racusin, 2014). In addition, men react to masculinity threats (feedback that they possess a “feminine” personality) defensively (Glick, Gangl, Gibb, Klumpner, & Weinberg, 2007), sometimes with aggression (Vandello, Bosson, Cohen, Burnaford, & Weaver, 2008). Thus, stereotypes of men (or at least White men; see Ghavami & Peplau, 2013) are problematic in that they are rigid and relatively unchanging, that while associated with positive characteristics such as power and status, they also have substantial negative overtones, and that the cultural perception of manhood as an earned status promotes a need to defend and protect it, often through aggression (Bosson, Vandello, Burnaford, Weaver, & Arzu Wasti, 2009). Redefining or broadening the perceived social role of men (and consequently stereotypes of men) may result in a more flexible, less negative impression of men. A less constrained definition of manhood and masculinity would likely afford greater flexibility in how to “be a man,” an outcome desirable not only for men, but as a means for decreasing rigid demarcations between the genders more broadly.

In contrast to stable stereotypes of men, fathers are perceived to have undergone substantial change in their trait characteristics. Using a paradigm modeled on Diekman and Eagly (2000), Banchefsky and Park (2016) asked participants about the perceived trait attributes of moms and dads in 1950, the present (2009), and in 2050. Their findings suggest that fathers were seen as having changed during this period, and were expected to continue to change into the future, with a greater likelihood that they possessed maternal traits (e.g., expressive, intuitive), and decreased likelihood that they possessed paternal traits (e.g., ambitious, assertive). Moreover, the magnitude of stereotype change was mediated by perceived change in the behaviors fathers were perceived as completing in the course of fulfilling their role as dads, with an increased likelihood of performing behaviors traditionally defined as part of the maternal role (e.g., comforting an upset child) and a decreased likelihood of paternal role behaviors (e.g., provide household income); evidence of this direction of mediation was greater than the reverse pattern. These findings raise the question of whether, as men take on the role of “new involved father” and its associated traits, this might promote change in perceptions of what men are like in general (Brandth & Kvande, 1998; Pleck, 1997; Pleck & Masciadrelli, 2004). Just as perceptions of women have changed as a result of new roles that they have taken on in the work domain, perceptions of men might similarly be changed in a beneficial manner as a result of highlighting their social role as fathers.

In the present research, we hypothesized that the trait attributes associated with fathers are in many ways distinct and different from those associated with men. To the extent, this is true, highlighting the social role of men as fathers might produce changes in perceptions of what men are like to effect both a weaker and more positive stereotype of men. We begin by examining the degree of similarity in the trait profiles seen as characteristic of men, relative to dads, moms, and women. We hypothesize that there is a greater disconnect between the social category men and the social role of dad, than that of women with the role of mom. In a second study, we test whether an experimental manipulation priming the role of father might result in changes to the trait characteristics viewed as typical of men relative to a control condition, bolstering our argument that highlighting their social role as fathers can produce change in stereotypic perceptions of what men as a group are like. In a final study, we demonstrate that highlighting men’s social role as fathers mitigates men’s opposition to policies benefiting low status others following a threat to their group-based identity as men.

Study 1

Some evidence exists that the social role of mother is more tightly connected to the category women than the role of father is to men. Utilizing a Go/No-Go task to assess implicit (i.e., fast and relatively unfiltered) category associations, Park, Smith, and Correll (2010) asked participants to quickly respond whenever an exemplar from either of two categories appeared. In one block, participants were watching for stimuli from either the category women (e.g., Rebecca, Susan, Amanda) or the category mom (e.g., mom, mother, momma). In another block, they searched for stimuli from either the category men (e.g., Michael, Jason, Kevin) or the category dad (e.g., dad, father, daddy). Performance on this task was significantly better on the mom + women block (i.e., it was an easier task) than on the dad + men block. These findings suggest a strong association between the categories women and mom—when people think women, they think mother. The category women appears to be more strongly associated and tightly connected with the role mother than the category men is with father. When perceivers think men, they activate a body of knowledge, a piece of which is father, but not nearly with the same associative connectivity as between women and mother.

While these data are suggestive of a more dissociative relationship for father and man than mother and woman, the task measures only category associations (e.g., mother goes with women) devoid of semantic content. Study 1 directly assesses beliefs about the trait content of the categories men and dads, as well as women and moms. The primary goal was to test whether the trait content of the role of dad is less congruent with the category men than the trait content for mom is with women. Specifically, we hypothesized that

Hypothesis 1 (H1): The correspondence in trait profiles for men and dads should show less overlap (i.e., lower correlations) than those for women and moms.

Hypothesis 2 (H2): This should be broadly true across multiple sample demographics (i.e., respondent gender, parental status, political conservatism, and age).

Hypothesis 3 (H3): Given expectations of greater involvement for the “new father,” in conjunction with the strong role of mother in defining the category women, the trait profile ratings for dads should be strongly aligned with those of moms and women, whereas trait ratings of the category men should show much weaker overlap with moms and women.

Hypothesis 4 (H4): Based on existing research, mean favorability ratings of traits viewed as characteristic of men were expected to be less positive than those characteristic of women, moms, and dads, whereas the traits viewed as uncharacteristic of these groups were expected to be more negative than traits that are uncharacteristic of men.

Method

Trait selection

A total of 145 positive and negative traits were selected from Williams and Best’s (1990) sex stereotypes Adjective Check List and supplemented with additional traits from Diekman and Eagly (2000), Ganong and Coleman (1995), and Troilo and Coleman (2008) to ensure sufficient coverage of both gender and parent stereotypes. Roughly equal numbers of positive and negative traits for each of the four categories (men, women, dads, moms) were selected, along with some traits that were not anticipated to be highly descriptive of any of the categories (see Table 1).

Table 1.

Full Set of Traits Rated for the Typical Man, Woman, Dad, or Mom.

Adaptable	Courageous	Generous	Messy	Showoff
Adventurous	Creative	Gentle	Moody	Shy
Affectionate	Critical	Good-natured	Muscular	Sociable
Aggressive	Cruel	Greedy	Nagging	Softhearted
Aloof	Cynical	Gullible	Nurturing	Stable
Ambitious	Daring	Harsh	Opinionated	Stern
Analytical	Decisive	Hectic	Outspoken	Stressed
Arrogant	Defensive	Helpful	Overworked	Strict
Artistic	Demanding	Helpless	Patient	Strong
Assertive	Dependable	High strung	Perfectionist	Submissive
Attentive	Dependent	Hostile	Playful	Supportive
Authoritative	Determined	Imaginative	Pleasant	Sympathetic
Autocratic	Devious	Impulsive	Pleasure seeking	Talkative
Beautiful	Devoted	Incompetent	Poised	Tasteful
Bitter	Disciplinarian	Independent	Preoccupied	Tense
Boastful	Dominant	Individualistic	Protective	Thrifty
Bossy	Egotistical	Industrious	Provider	Trustworthy
Bumbling	Emotional	Insecure	Prudish	Unavailable
Calm	Energetic	Intelligent	Pushy	Understanding
Capable	Expressive	Intimidating	Rational	Unemotional
Careless	Fair minded	Intuitive	Reckless	Unfaithful
Charming	Fickle	Kind	Reliable	Uninvolved
Clear thinking	Flirtatious	Lazy	Rugged	Unprincipled
Coarse	Forceful	Lenient	Selfish	Untrustworthy
Competitive	Forgiving	Loud	Selfless	Vain
Complaining	Frank	Loyal	Self-reliant	Vindictive
Conceited	Frivolous	Mathematical	Sentimental	Warm
Confident	Funny	Mature	Sexy	Weak
Considerate	Fussy	Meek	Shallow	Whiny

Participants

Participants were recruited through Amazon’s Mechanical Turk marketplace, restricted to U.S. workers, and compensated US$0.50. Our goal was to recruit approximately 80 participants to rate each target group and 80 to rate trait favorability. Although there is no consensus concerning how to conduct power estimations for the type of analysis involved in Study 1, we reasoned that these numbers should result in stable mean estimates, and also permit us to examine variation in perceptions as a function of key demographic characteristics. Initially, 399 respondents completed the survey. Eleven of these (3%) were eliminated for failure to pass attention checks (described in the “Procedures” section), resulting in a final sample of 388 respondents (241 females, M_age = 33, SD = 12.17, median = 28, 74% White). Participants were randomly assigned to rate either men (n = 76), women (n = 76), dads (n = 79), moms (n = 81), or the favorability (n = 76) of the traits.

Procedures

To begin, participants were asked to think and briefly write about their assigned group (e.g., “think about men in the United States”). Next, they rated the extent to which each of the 145 traits (presented in a random order) described the “typical or average” group member (1 = not at all, 7 = extremely so). Six additional items were embedded within the traits to serve as attention checks. These were items that should be viewed as descriptive of all humans and therefore of all four groups (alive, living, person, human, mortal, breathing). Subjects who rated the group below the scale midpoint (4) on four or more of these attention checks were dropped from analyses (n = 6). Participants assigned to the favorability condition were asked to rate the extent to which each trait was a favorable (7 = very favorable) or unfavorable (1 = very unfavorable) quality. For attention checks, eight highly positive and negative traits were examined. If a participant rated 4 or more of either type incorrectly (e.g., cruel was rated as favorable, or loyal was rated as unfavorable), they were flagged as inattentive and dropped (n = 5). Participants also provided their gender, age, parental status, and self-rated political orientation (1 = extremely liberal; 7 = extremely conservative; M = 3.52, SD = 1.69). These particular demographics were selected because each was deemed a plausible moderator of H1, and accordingly used to examine H2, the generality of the predicted greater overlap for women and moms than men and dads. Political conservatism has been associated with resistance to change and the need to manage uncertainty (Jost, Glaser, Kruglanski, & Sulloway, 2003), and thus knowing whether the predicted pattern was independent of political conservatism was conceptually useful. Note that the data for moms and dads were used in Study 1 of Hodges and Park (2013), to examine overlap in trait ratings of professionals with that of moms and dads.

Results

The mean rating on each trait for each target group was calculated across raters. The correlations in these mean ratings were then calculated for women and moms, and also for men and dads, as an index of the degree of overlap in the perceived trait profiles. All of the reported analyses for Study 1 use trait as the unit of analysis. As seen in Table 2, the correlations show that in support of H1, the average trait profiles for women and moms closely mirrored each other (.87), more so than those for men and dads (.56, z = 6.13, p < .001, based on Raghunathan, Rosenthal, & Rubin’s, 1996, z test for nonoverlapping dependent correlations; the sample is nonoverlapping because the compared correlations of moms versus women and dads versus men do not overlap in terms of the object being rated; and it is dependent because the traits compared in each correlation are the same).

Table 2.

Correlation of Mean Trait Profiles for Moms and Women Versus Dads and Men Across All Respondents and Within Various Demographic Subgroups (Study 1).

Participant group	Moms and women		Dads and men		z test of difference	Effect size
Participant group	r	95% CI	r	95% CI	z test of difference	Effect size
All participants (n = 312)	.87**	[0.87, 1]	.56**	[0.47, 0.80]	6.13**	.70
Females (n = 194)	.89**	[0.89, 1]	.46**	[0.33, 0.46]	7.67**	.93
Males (n = 118)	.80**	[0.80, 1]	.72**	[0.72, 1]	1.60^†	.19
Parents (n = 125)	.89**	[0.89, 1]	.65**	[0.61, 0.94]	5.96**	.65
Nonparents (n = 187)	.83*	[0.83, 1]	.54*	[0.44, 0.77]	4.90**	.58
Conservatives (n = 96)	.89**	[0.89, 1]	.65**	[0.61, 0.94]	5.60**	.65
Liberals (n = 149)	.83**	[0.83, 1]	.58**	[0.50, 0.83]	4.67**	.53
Age 18-25 (n = 87)	.73**	[0.73, 1]	.45**	[0.32, 0.65]	3.67**	.44
Age 26-35 (n = 81)	.86**	[0.86, 1]	.52**	[0.41, 0.74]	6.21**	.72
Age 36+ (n = 89)	.82**	[0.82, 1]	.53**	[0.43, 0.75]	5.25**	.57

Note. The number of observations for each correlation is the number of traits (145). The number of participants on which each mean trait rating is based appears as n for that group. Liberals = self-ratings of 1 to 3 and conservatives = 5 to 7 on the political orientation scale. The difference in the correlations was tested using the z test for nonoverlapping dependent correlations (Raghunathan et al., 1996). The effect size is based on Cohen’s q, estimating the effect size for the difference between two correlations with the following guidelines for interpretation: <.1: no effect; .1 to .3: small effect; .3 to .5: intermediate effect; >.5: large effect (https://www.psychometrica.de/effect_size.html#cohenq). CI = confidence interval.

†

p = .10. *p < .05. **p < .01.

The mean trait profiles were also computed breaking the sample down based on key demographic characteristics (see Table 2). The pattern of greater overlap for moms and women than dads and men was remarkably consistent across these various subgroups (H2). Both male and female participants shared this view, although the difference was decidedly larger in magnitude for females, and only marginally significant for males (p = .10). The pattern held for parents and nonparents (respondents who were stepparents but not biological parents were included in the parent data), for conservatives and liberals, and for all three age cohorts. Note that the participant numbers become very small for some cells if the ratings are simultaneously divided on more than one subgroup.

Another way to characterize the degree of overlap in perceptions is by asking how frequently the same versus different traits were used to characterize the groups. Within each group, the raw ratings of participants were used to calculate a t test for each trait, asking whether the group was rated as significantly above or below the scale midpoint (i.e., the group was viewed as possessing that trait, or decidedly lacking that trait). Many of the traits were seen as either descriptive of all four groups (35 out of 145, or 24%; see Figure 1), or not descriptive of any of the four groups (32, or 22%; e.g., lenient, hostile, shy). Of the remaining traits, only eight (6%) uniquely characterized men and dads and no other group, whereas 13 (9%) uniquely characterized women and moms, and 17 (12%) were shared by moms, dads, and women, but not men. And though only five traits (3%) uniquely characterized women and no other group, 21 traits (14%) were only attributed to men. Moms had just one unique trait and dads had zero.

Figure 1.

Shared overlap in traits viewed as characteristic of moms, dads, women, and men (Study 1).

According to H3, given expectations of greater involvement for the “new father,” in conjunction with the strong role of mother in defining the category women, the trait profile ratings for these three categories were predicted to be strongly aligned, whereas the trait profile of men was expected to be relatively distinct. Support for H3 can be seen both in Figure 1 (“Shared by Three Groups” section) and in correlations of the trait profiles for the groups. The mean trait profile for dads was correlated .83 with that for moms, and .63 for that with women (recall it correlated only .56 with men). In contrast, the trait profile for men was correlated .18 (ps < .03) with both moms and women. Although still significant, the latter correlations clearly were of a much smaller magnitude than any of the other comparisons.

Consistent with H4, many of the traits seen as uniquely characteristic of men were strikingly negative in tone (see Figure 1). As seen in Figure 2, the average favorability ratings of the traits that men were seen to possess were lower than traits viewed as characteristic of the other three groups (ts comparing men with each of the other three groups all >4.16, smallest M_difference = .93, 95% confidence interval [CI] = [0.49, 1.37]; all ps < .001; all ds > .65) and those they lacked were relatively more favorable than those lacked by the other three groups (all ts > 3.72, smallest M_difference = 1.21, 95% CI = [0.56, 1.85]; all ps < .001; all ds > .81). For moms, dads, and women, the mean favorability ratings of traits they were judged to possess were all significantly above the scale midpoint of 4, all ts > 8.97, lowest M = 5.19 (95% CI = [4.93, 5.46]); all ps < .001; all ds > .16, whereas the traits they were judged to lack were below the scale midpoint, all ts > 8.66, highest M = 2.47 (95% CI = [2.11, 3.83]); all ps < .001; all ds > 1.61. For men, the mean favorability of the traits possessed was not above the scale midpoint, t(76) = 1.47, M = 4.26 (95% CI = [3.91, 4.62]), p = .14, nor was the favorability of the traits lacked below the scale midpoint, t(36) = 1.18, M = 3.68, (95% CI = [3.13, 4.23]), p = .25. In fact, the difference in favorability for the traits possessed versus lacked by men was only marginally significant, t(143) = 1.71, M_difference = 0.56 (95% CI = [–0.09, 1.21]), p = .09, and as is clear in Figure 2, much smaller in magnitude than for the other three groups.

Figure 2.

Mean favorability of traits rated as characteristic (possess) and uncharacteristic (lack) of moms, dads, women, and men (Study 1).

Discussion

While the trait content of the category women was largely redundant with that of mothers, the category men corresponded much less closely to that of fathers, and this was true across multiple demographic subgroups, including gender (albeit marginally for male participants), age, political orientation, and parent status. Based on the trait profiles emerging from Study 1, current societal definitions of men and father share relatively less in common than women and mother, and indeed, father shows as great of overlap with women (r = .63) as with men (r = .56). Moreover, the profile of traits uniquely ascribed to men was quite negative in tone, particularly compared with the other three groups. In developing the Masculine Norms Inventory, Mahalik et al. (2003) settled on 11 distinct factors, a surprising number of which were also quite negative in tone (e.g., violence, dominance, playboy, power over women, disdain for homosexuals). Not only does it appear that women (and moms and dads) are rated as wonderful (Eagly, Mladinic, & Otto, 1991) but also men are rated as decidedly not wonderful (Eagly & Mladinic, 1989; Prentice & Carranza, 2002).

Study 2

Study 2 examined whether the social role fathers could be leveraged to shift stereotypes surrounding men. In considering what men are like, if social perceivers first bring to mind their social role as fathers, could this alter the static and restrictive characterization of men seen in previous research (Croft et al., 2015; Diekman & Eagly, 2000; Twenge, 1997) and the relatively unflattering characterization observed in Study 1? Study 2 tested this hypothesis by having participants first read about factual current trends within the United States prior to making trait ratings of men as a group. In one condition, the trends emphasized the social role of men as fathers. In the other, intended to serve as an active control condition, the trends described characteristics of the American workforce. We hypothesized that

Hypothesis 5 (H5): Ratings of men in the workforce prime condition would closely parallel those of men (and not of dads) from Study 1, whereas in the father prime condition, ratings of men would map more closely onto those of dads from Study 1.

Hypothesis 6 (H6): As a result, the trait profile of men in the father prime condition was predicted to be decidedly more positive than that in the workforce prime condition.

Method

Priming condition materials

Two reports from the Pew Research Center were used to develop priming materials for the two conditions. One described current trends in the American workforce (http://www.pewresearch.org/fact-tank/2016/10/06/key-findings-about-the-american-workforce-and-the-changing-job-market/), and the other described current trends in fatherhood (http://www.pewresearch.org/fact-tank/2016/06/16/fathers-day-facts/). The reports were used to create a seven-panel slide show for each condition, with an introductory slide, and then six fact slides, each with a graphic depiction and text description of facts taken from the Pew reports. See Figure 3 for an example slide from each condition. In general, the facts in the father prime condition emphasized the changing role of fathers, including less responsibility for solely supporting the family financially and greater involvement in child care, and the centrality of being a father to their identity. The workforce condition discussed greater job opportunities for skilled workers and the importance of training, both in the present and over the work-life, for maintaining one’s place in the workforce. Each slide was presented for a minimum time window, and participants could not advance until this window had passed.

Figure 3.

Example facts from the father prime (top panel) and workforce prime (bottom panel) conditions (Study 2).

Participants

Participants were recruited through Amazon’s Mechanical Turk marketplace, restricted to U.S. workers, and compensated US$0.75. Initially, 151 respondents completed the survey. Four of these (3%) were eliminated for failure to pass attention checks (using the same criteria as Study 1), and one was eliminated for rating all the traits a 4, making it impossible to calculate trait profile correlations, resulting in a final sample of 146 respondents (78 females, M_age = 37, SD = 12.28, median = 34, 84% White). Participants were randomly assigned either to the father prime (n = 76) or workforce prime (n = 70) conditions. A power analysis using G*Power and assuming medium effect sizes (f = .25) indicated that our planned sample of 150 participants provided 99% power to detect an interaction between prime and target (H5), and 86% power to detect an effect of prime on favorability (H6; see Richard, Bond, & Stokes-Zoota, 2003, whose meta-analysis of meta-analyses showed that the typical effect size is d = .45).

Procedures

Participants were told they would be making ratings of the degree to which various traits characterize groups and that “we have found that it helps in making these trait judgments to first think about current trends within the United States that are affecting people’s lives.” This was followed by either the father prime or workforce prime slide show, after which they were asked to summarize the content of the research findings presented in the slide show, and to state what they thought of the trends and whether they agreed with the claims. Next all participants were told they would be rating “men in the United States.” As in Study 1, they began by thinking and briefly writing about the group, and then rated the extent to which the same 145 traits plus six attention check items from Study 1 described men in the United States (1 = not at all, 7 = extremely so). Participants also provided their gender, age, parental status, and self-rated political orientation (1 = extremely liberal, 7 = extremely conservative; M = 3.54, SD = 1.8).

Results

Participants’ trait ratings of men as a function of prime condition were compared with the trait profiles (the mean trait ratings calculated across raters) for men and for dads from Study 1. Specifically, for each participant, two correlations were computed using his or her ratings of the 145 traits, one examining correspondence with the average trait profile for men from Study 1, and one with that for dads from Study 1. The resulting correlations were subjected to a Fisher’s z transformation and analyzed in a 2 (prime condition: workforce vs. father) × 2 (target: correlation with Study 1 trait profile for men vs. dads) × 2 (subject gender) ANOVA with repeated measures on the second factor, target.

The predicted Prime × Target interaction was significant, F(1, 142) = 6.60, b = 0.07 (95% CI = [0.02, 0.13]), p = .011, $η_{p}^{2}$ = .05. Consistent with H5, trait ratings of men following the father prime mapped more closely on to those of dads from Study 1 than did ratings of men following the workforce prime (the leftmost bars in Figure 4; note the r values are plotted for ease of interpretation), F(1, 142) = 7.41, b = 0.08 (95% CI = [0.02, 0.14]), p = .007, $η_{p}^{2}$ = .05. Interestingly, the correspondence of the trait profile ratings of men in Study 2 with the trait profile of men from Study 1 was equally high in the father and workforce prime conditions (see Figure 4), F < 1. Thus, thinking about current trends surrounding fatherhood in the United States, relative to thinking about workforce trends, resulted in trait ratings of men that mapped more closely onto those of dads from Study 1, but were no less correspondent with trait ratings of men from Study 1. This suggests that increasing the salience of men’s social roles as fathers might very well result in a broader and less restrictive profile of how men are seen, adding traits seen as characteristic of fathers, but not at the expense of traits seen as characteristic of men (i.e., at no cost to “manhood”). Both the target and prime main effects were also significant although subsumed under the predicted interaction reported above (both main effects can be seen in Figure 4). Average correlations with the men profile were greater than those with dads, F(1, 142) = 8.40, p = .004, and those in the fatherhood prime condition were larger on average than those in the workforce prime, F(1, 142) = 4.35, p = .04. Finally, participant gender interacted with target, F(1, 142) = 4.33, p = .04, such that although overlap with the dad profile did not depend on gender, F < 1, on average female participants rated men more similarly to the men profile from Study 1 than did male participants, F(1, 142) = 7.24, p = .008. Importantly, the predicted prime by target interaction (H5) was not qualified by participant gender, F(1, 142) = 1.13, p = .29. No other effects were significant, all Fs < 1.64, ps > .20.

Figure 4.

Average correlation of trait ratings for men in Study 2 by prime condition with the average trait profiles for dads, men, women, and moms from Study 1.

The mean trait ratings of women from Study 1 were similarly used to examine overlap with the trait ratings of men from Study 2. Notably, this overlap was significantly greater in the father prime than workforce prime condition, F(1, 142) = 8.36, b = 0.07 (95% CI = [0.02, 0.12]), p = .005, $η_{p}^{2}$ = .06 (see Figure 4). Thus, priming men’s role as fathers resulted not only in a trait profile for men that more nearly mapped on to that of dads but also in perceptions of men as more similar to (or less different from) women, suggesting that increasing the tie between men and their role as fathers might decrease divisions in how the genders are seen. Finally, trait ratings of men in Study 2 also showed greater overlap with that of moms from Study 1 in the fatherhood than workforce prime condition, F(1, 142) = 9.47, b = 0.09 (95% CI = [0.03, 0.15]), p = .003, $η_{p}^{2}$ = .06 (see Figure 4). No other significant effects emerged, all Fs < 1.

H6 suggested that as a consequence of this shift in the characterization of men, the average favorability of traits men were judged to possess would be more positive following the father prime, and the traits men were judged to not possess more negative, relative to the workforce prime condition. As in Study 1, within condition, the raw ratings of participants were used to calculate a t test for each trait dimension, asking whether men were rated as significantly above the scale midpoint on that trait (men were judged to possess that trait) or below the scale midpoint (men were judged to lack that trait). As is clear in Figure 5, support for H6 was obtained (tests of H6 use trait as the unit of analysis). In both conditions, the traits men were viewed to possess were more favorable than those they were seen to lack, father prime: t(104) = 8.58, b = 1.14 (95% CI = [0.88, 1.40]), p < .001, $η_{p}^{2}$ = .41; workforce prime: t(97) = 2.39, b = 0.42 (95% CI = [0.07, 0.76]), p = .02, $η_{p}^{2}$ = .05, but the magnitude of this difference was much larger in the father prime than workforce prime condition, F(1, 201) = 11.08, b = 0.36 (95% CI = [0.15, .57]), p = .001, $η_{p}^{2}$ = .05, for the Prime × Possess/Lack interaction. Testing the simple effect of prime condition within possess versus lack, the traits seen as characteristic of men were more favorable in the father prime than workforce prime condition, t(138) = 2.18, b = 0.27 (95% CI = [0.02, 0.52]), p = .03, $η_{p}^{2}$ = .03, and those seen as uncharacteristic of men were more negative in the father prime than workforce prime condition, t(63) = −2.69, b = –.45 (95% CI = [–0.78, –0.12]), p = .009, $η_{p}^{2}$ = .10. Table 3 lists all the traits for which men were rated significantly different as a function of prime condition. It is clear from the table that a much more favorable portrayal of men was elicited when the social role of fatherhood was first brought to mind than when the role of worker was considered.

Figure 5.

Mean favorability of traits rated as characteristic (possess) and uncharacteristic (lack) of men by condition (Study 2).

Table 3.

Traits Rated as More or Less Characteristic of Men as a Function of Father Versus Workforce Prime (Study 2).

When participants first read and thought about current trends in Fatherhood, relative to the Workforce, Men were rated as significantly:
More . . .		and Less . . .
Attentive	Loyal	Aloof	Greedy
Dependable	Muscular	Arrogant	Hostile
Devoted	Nurturing	Bitter	Loud
Energetic	Selfless	Boastful	Pushy
Generous	Softhearted	Coarse	Selfish
Gentle	Stable	Conceited	Stressed
Helpful	Supportive	Cruel	Uninvolved
		Cynical	Unprincipled
		Demanding	Untrustworthy
		Devious	Vain
		Dominant	Whiny

Discussion

The results from Study 2 suggest that one means for shifting rigid and relatively negative stereotypes of men is by highlighting their social role as fathers. Participants who first read about current trends in fatherhood rated men more similarly to the average trait profile for dads from Study 1 than did those who read about current trends in the workforce. Of note, thinking about men in their role as fathers did not result in perceptions that corresponded any less to those of men from Study 1 relative to thinking about their role in the workforce. That is, priming the social role father appeared to increase correspondence with the view of men as dads at no cost to the view of men as men. This would seem to make this approach more broadly attractive in that the message is not that men need to eschew their identities as men, but rather to embrace their identities or potential identities as fathers. The finding that men were also seen more similarly to women following the father prime suggests that highlighting their role as fathers might additionally help to dissolve rigid demarcations between the genders. Finally, thinking about men in their role as fathers resulted in a more positive, and less negative, profile of trait ratings of men than thinking about their role in the workforce.

Study 3

The findings from Studies 1 and 2 indicate that while women are viewed very similarly to moms, men are viewed as distinct and different from dads, and the trait profile of men is much more negative than that of dads. Highlighting the social role of men as dads increased the degree of correspondence in the trait profile of men-to-dads and reduced the negativity of traits associated with men. These findings raise the question of whether downstream consequences might emerge on social judgments as a function of highlighting the social role of men as dads. Specifically, past research has shown that a group-based realistic threat to men results in reactivity against policies that benefit lower status outgroups (Renfro, Duran, Stephan, & Clason, 2006; Willer, Rogalin, Conlon, & Wojnowicz, 2013), and increase aggressive responses (Bosson et al., 2009). We wondered whether highlighting the social role of men as fathers might serve to mitigate that threat and decrease negative reactivity. This was the primary goal of Study 3. Male participants began by reading a real op-ed piece focused on current problematic social trends surrounding men in the United States, or a control article. Following this threat, they were exposed to the same father prime or workforce prime manipulations as in Study 2. Next they rated their support for various social policies beneficial to lower status outgroups (e.g., immigrants, women, homosexuals), or aggression-related policies (support for the death penalty). We hypothesized that the threat manipulation would result in greater opposition to benevolent policies and increased support for punitive policies, but that exposure to the father prime would mitigate this threat reaction.

Furthermore, this hypothesized shift in policy support might occur as a group-based process, or it might be mediated through shifts in self-perception. Specifically, the initial threat might activate concerns about what men as a group are like, but bringing to mind men’s social role as fathers serves to mitigate this threat reaction, in turn decreasing opposition to benevolent social policies. Alternatively, shifts in policy support might occur through a process of redefinition of the self. In this case, participants in the threat, father prime condition, may redefine the self to be more in line with the trait profile of dads, and this redefinition is what decreases opposition to benevolent social policies. Study 3 tested these hypotheses with male participants.

Hypothesis 7 (H7): Exposure to an opinion piece threatening to men should result in opposition to benevolent social policies, but that opposition should be mitigated following exposure to the father prime, relative to the workforce prime.

Hypothesis 8 (H8): Threatened men exposed to the father prime may demonstrate greater self-dad than self-men overlap in trait ratings, relative to those in the workforce prime condition, and such changes in self-definition may then mediate the hypothesized condition differences on benevolent policy support.

Method

Participants

Participants were recruited through Amazon’s Mechanical Turk marketplace, restricted to U.S. workers, and compensated US$2. To recruit only men, initial instructions stated that enough women had completed the study and to continue only if the worker was a man. At the end of the survey, we noted that it was critical in terms of the integrity of the data to have gender correctly recorded, and that if the worker was a female we would still pay them and add them to our female sample. Initially 249 participants completed the study. Of those, 16 acknowledged they were not male. In addition, three gave ratings of 4 to every item; four gave unsatisfactory responses to the open-ended queries following the opinion pieces (e.g., no response or “none”); 14 failed the attention check item in the policies (see below); and one subject was identified as an outlier on the policy ratings (the only subject with a studentized residual greater than 3). This resulted in a final dataset of 211 males, M_age = 35, SD = 10.09, median = 33, 78% White. Participants were randomly assigned to condition: threat + father prime (n = 50), threat + workforce prime (n = 52), no threat + father prime (n = 56), and no threat + workforce prime (n = 53). A power analysis using G*Power and assuming medium effect sizes (f = .25) indicated that our planned sample of 250 participants provided 98% power, and our obtained sample 95.10% power, to detect an interaction between threat and prime. As in Study 1, there is no consensus on how to test statistical power for the analyses used to test H8. Roughly, 52 people per cell seemed sufficient to test this hypothesis.

Materials

Opinion pieces

Two op-ed pieces were taken from The New York Times and modified in part to decrease length. The threat piece (retitled, “Is Our Biggest Social Problem Men?” Chira, 2016) discussed problematic societal trends surrounding men, including lower rates of college completion, high unemployment rates, drug use, high levels of health problems, and high crime rates. The control editorial (“Google Wants Driverless Cars, But Do We?” Kitman, 2016) discussed concerns around driverless cars, including unsubstantiated claims regarding greater safety, security concerns with cars being hacked, and massive infrastructure upgrades that will be needed.

Priming manipulation

The father and workforce prime materials were identical to those used in Study 2.

Policies

Seven policy items were taken or adapted from the American National Election Survey 2016 Pilot Study (ANES; 2016). These included opposition to immigration (two items), opposition to policies that benefit women in the workforce (equal pay; holding jobs for 6 months following the birth or adoption of a new child), opposition to increased minority student representation at universities through admissions policies, opposition to homosexuals serving openly in the military, and opposition to the death penalty in murder convictions (reverse coded). All responses were on 7-point scales using the ANES labels (e.g., 1 = favor a great deal; 7 = oppose a great deal).

Procedure

The purpose of the study was described as examining reactions to different types of media. Participants were told they would read two articles, an opinion piece that takes a position on a current trend, and a fact-based article describing a current trend, both of which were published in 2016. We noted that multiple articles of each type were being studied but that each participant would read just one of each. They were asked to first read the opinion piece and to approach it as they might typically read an opinion piece. They were required to spend 2 min on the page before advancing. On the next screen, they were asked to summarize the piece in 2 to 3 sentences and to say a bit about their reactions. Next, they were told they would read the fact-based article. The instructions and procedure for this were identical to the father and workforce primes in Study 2. Next, participants were told that the research team could better understand their reactions to the articles by knowing more about them as a person. They rated the 145 traits from Studies 1 and 2 in terms of how much each characterizes or describes them (1 = not at all; 7 = extremely so). Finally, they were asked to indicate where they stood on “the following attitude and policy issues that are often debated in online media.”

Results

Policy opposition

Endorsement of the seven policy items reflecting opposition to immigrants, women, and gays, and support for the death penalty were strongly related to one another (α = .76). Accordingly, scores on these were averaged into a single measure termed opposition to benevolent policies. This measure was analyzed in a 2 × 2 ANOVA as a function of threat condition (threat vs. no threat) and prime (father vs. workforce). H7 predicted that a threat to the group men should result in opposition to benevolent social policies, but that considering the social role of fatherhood should serve to mitigate this threat, thereby decreasing opposition. This predicted interaction between threat condition and prime was in fact significant, F(1, 207) = 4.00, b = 0.16 (95% CI = [0.002, 0.32]), p = .05, $η_{p}^{2}$ = .02 (see Figure 6). In the threat condition, opposition was significantly lower following the father prime than after the workforce prime, F(1, 207) = 3.97, b = 0.23 (95% CI = [0.002, 0.45]), p = .05, $η_{p}^{2}$ = .04, whereas prime condition did not affect mean opposition levels in the no threat condition, F(1, 207) = .67, b = 0.09 (95% CI = [–0.13, 0.31]), p = .41, $η_{p}^{2}$ = .01. Neither of the main effects was significant, both Fs < 1. The pattern of the means was as predicted, with the threat + workforce prime condition demonstrating the greatest opposition, the threat + father prime the lowest, and the two no threat conditions falling in between these.

Figure 6.

Mean opposition to policies that benefit oppressed groups, and support for the death penalty, as a function of threat condition and prime (Study 3).

Self-trait ratings

The father prime condition may have led to more support for benevolent social policies because it highlighted an alternative, more positive characterization of what men as a group are like, thus diminishing the threat. Alternatively, the protective mechanism may have operated more specifically through a change in self-perceptions. That is, participants may have shifted their self-definitions to bring them more in line with the trait profile dads, and this may have mitigated the threat, reducing opposition to the policies (H8). To test this hypothesis, for each participant, his self-ratings on the 145 traits were correlated with the trait profile ratings for both men and dads from Study 1. These correlations were z transformed and analyzed as a function of threat and prime condition, treating target (dad vs. men) as a within-participant factor. Of note, on average participants rated themselves much more strongly in line with the trait profile of dads than men (average r = .54 vs. .19), F(1, 207) = 732.57, b = 0.44 (95% CI = [0.41, 0.48]), p < .001. However, this difference did not depend on condition, all Fs <1, see Figure 7. The Threat × Prime interaction across target was significant such that participants in the threat + workforce prime and no threat + father prime conditions rated themselves more similarly to both dads and men compared with the other two conditions, F(1, 207) = 4.08, p = .04. This interaction was not predicted and did not depend on target, which was the conceptual question of interest (that the conditions would differentially affect self-overlap with dads relative to men).

Figure 7.

Average correlation between self-ratings in Study 3 and the mean trait profile of dads and men from Study 1, by threat condition and prime.

The finding that participants rated themselves more similarly to dads than men was surprising, but in retrospect, it likely reflects the quite negative trait profile of men, which is unlikely to be endorsed for the self. As evidence that the self-trait ratings behaved reasonably, participants who were fathers rated themselves as particularly similar to dads (r = .62, n = 75) relative to those who were not fathers (r = .49, n = 136), F(1, 203) = 19.98, p < .001. Thus, although ratings of the self appeared reasonable on several levels (greater overlap with the more positive dads profile; particularly greater overlap with the dads profile for participants who were themselves fathers), given the lack of condition differences in the magnitude of self-to-dad relative to self-to-men correlations, no evidence emerged to suggest that changes to policy endorsement were driven by changes in self-perceptions.

Discussion

The goal of Study 3 was to examine whether priming men’s social role as fathers could result in reparative downstream consequences. Specifically, past research has demonstrated greater opposition to social policies that benefit low status groups for those who experience a threat to the social category men (Renfro et al., 2006; Willer et al., 2013). We examined whether such effects might be mitigated by consideration of men’s social role as fathers (H7). We did indeed see evidence in support of this hypothesis. Following a realistic group-based threat to men (that men as a group are failing in society), male participants who then read about the increased role of American fathers in their children’s lives demonstrated significantly less opposition to social policies that would benefit lower status groups (immigrants, women, ethnic minority groups, homosexuals), and less support for the death penalty, than those who read a control prime discussing current trends in the American workforce. No such difference emerged for those who did not experience the initial threat to the social category men.

Of note, however, this effect did not appear to operate through a change in self-perceptions (no support for H8). Although self-ratings on the traits behaved reasonably (e.g., greater self-overlap with the more positive dads than men trait profile; greater self-overlap with the dads trait profile for participants who were themselves fathers), the manipulations did not affect self-trait-perceptions. This finding is consistent with other research showing that intergroup dynamics are often driven by group, rather than personal level perceptions (Crosby, 1984; Stephan & Renfro, 2002). Of note, work by Spears, Doosje, and Ellemers (1997) suggests that it might be difficult to see changes in self-perceptions at the level of trait judgments, but that a more global and abstract judgment such as “how similar are you to the typical adult American male” might show effects (perhaps in conjunction with preexisting differences in level of identification). Future research that tests whether the manipulation used here could change a more global self-judgment would be useful. Similarly, the current manipulation utilized a group (American men) as opposed to individual-level threat typically seen in work on precarious manhood (your personal level of masculinity; Bosson et al., 2009; Vandello et al., 2008). Perhaps an individual-level threat would be more likely to affect self-trait-perceptions, although whether such a threat would also produce changes in support for group-based social policies is unclear. For now, the current study shows that the effect of a group-based realistic threat on opposition to benevolent social policies can be mitigated by consideration of a key social role men occupy, that of fathers.

General Discussion

Although stereotypes of women have undergone significant change as a function of women taking on a greater number and variety of social roles in the workforce, stereotypes of men have not shown a parallel evolution. Men as a group are perceived to have retained roughly the same high level of masculine, and the same low level of feminine, trait characteristics over time, and those traits are expected to persist into the future (Diekman & Eagly, 2000). Moreover, the traits ascribed to men tend to be relatively more negative than those ascribed to women (Eagly & Mladinic, 1989; Prentice & Carranza, 2002).

Motivated by findings of Banchefsky and Park (2016), who demonstrated that, in contrast to relatively stable perceptions of men, fathers were perceived to have changed over the past 50 years and were expected to continue to change into the future, we hypothesized that one means for bringing change to stereotypes of men might be to highlight their social role as fathers. Indeed, portrayals of “the new involved father” depict fathers as taking on a more involved and nurturing role in raising their children (Brandth & Kvande, 1998; Pleck, 1997; Pleck & Masciadrelli, 2004). Study 1 demonstrated that the traits perceived as characterizing men are more distinct and different from dads, than the traits seen as characteristic of women are from moms. In fact, women, moms, and dads were seen as possessing many of the same trait attributes, resulting in a high degree of overlap in the trait profiles of these three groups. That men were seen less similarly to dads was true across a number of different subpopulations, including participants with and without children, political liberals and conservatives, and different age groups. Male and female participants both expressed more similar views of women and moms than men and dads, although the effect was of a much smaller magnitude and only marginally significant for male participants. Importantly, men were judged to possess traits that were less positive in content, and to lack traits that were more positive in content, relative to the other three groups.

In Study 2, we hypothesized that just as stereotypes of women have changed to include more agentic traits as a function of participation in the workforce (Eagly & Wood, 2012), stereotypes of men might similarly be altered to include traits typically associated with fathers by highlighting men’s social role as fathers. Indeed, participants who first read and thought about current trends in fatherhood rated men in a manner that more nearly paralleled that of dads from Study 1 relative to those who considered current trends in the workforce, while still rating them as similarly to men from Study 1 as those in the workforce prime condition. Moreover, the trait profile of men in the father prime condition was more positive than that in the workforce prime condition. Finally, Study 3 demonstrated a possible downstream consequence for men of bringing to mind their social role as fathers. Following a threat to the group men, men for whom the role of father was made salient expressed greater acceptance of social policies beneficial to lower status groups than those who read the control, workforce prime. To be clear, we certainly are not arguing that such a short priming manipulation as that used in Studies 2 and 3 would have lasting effects, but we would argue that if the social role of men as fathers was persistently highlighted in the media, in movies, and in the lives of public figures, this could result in profound and lasting change in terms of what comes to mind when one thinks of men as a social category.

One strength of the current research is that it utilized a large number of diverse traits (145) that, importantly, varied in favorability. Often researchers use just a small number of traits, and typically these are all positive in tone. For example, in a recent study by Haines, Deaux, and Lofaro (2016), they argue that stereotypes of men and women have remained relatively unchanged over the past 30 years. However, their measures included just 16 traits (eight communal and eight agentic), all of which were positive in tone (to their credit, Haines et al. measured other dimensions including roles, occupations, and physical characteristics, which is, in part, why so few traits were assessed). The findings in the current research indicate the importance of measuring both positive and negative traits to assess perceived change both with respect to stereotype content as well as change in the valence of the group attitude.

The pattern of results from the current studies supports the argument by Koenig and Eagly (2014) that one means of changing future expected stereotypes is through pairing groups with social roles that differ from their current roles. They discuss possible barriers to entry into new roles, for example, role incongruity theory and backlash effects. In this case, it is not so much that men would take on a new role—by definition, men are the ones who take on the role of father. The manipulation did not describe men taking on a “new” social role but rather called attention to current trends in how they are fulfilling an existing one, and tying that role more closely to the group. The current studies therefore suggest that group stereotypes can change not only from shifting roles, but by shifting the behaviors associated with a current social role (e.g., the “new involved father”).

Consistent with the asymmetry in the strength of connection to parent roles for the genders, other research has demonstrated that the category moms is viewed in highly essentialist terms, more so than the category dads (Park, Banchefsky, & Reynolds, 2015). That is, mothers are viewed as a deep and meaningful category, and membership in the category has significant ramifications for what one’s life is like; it is a permanent and stable group membership. This is less true for fathers. One means for increasing the tie between the social role father and the category men might reasonably be to find ways to increase the perceived essentialism of fathers (McPherson, Banchefsky, & Park, in press). If becoming a father is similarly viewed as evoking a permanent, far-reaching, influential “real” change, this in principle may lead to a corresponding change in the trait characteristics associated with men, who again, by definition, are the group that occupies the social role of father.

There are important limitations to this work that should be acknowledged. First, work by Ghavami and Peplau (2013) suggests that when people think about the category men, they are primarily thinking about White men. In the current studies where race was left unspecified, perceivers most likely brought to mind White men and women, or White dads and moms in making their responses. It is important to explore the processes studied here as they explicitly apply to non-White groups. Indeed, one wonders whether, given the often intensely negative stereotype of Black men, highlighting passionately involved Black fathers could be a pathway for powerful change in the content and negativity of this stereotype. Also, the participants included in these studies represent a particular sample (MTurk workers). While H2 explored some individual difference variables, there are certainly others that could be important in moderating the effects that were not examined here.

With time, increasing the tie between men’s social role as fathers should help to redefine the stereotype of men and conceptions of masculinity more broadly (Eagly & Wood, 2012), potentially freeing even those who are not fathers from the strictures of current societal definitions of manhood. Such an outcome would seemingly be a win all around. Men would be at least somewhat freed from the burden of fighting to maintain precarious manhood, and families would be strengthened through better overlapping goals of mothers and fathers. Importantly, by increasing the salience of their roles as parents for both women and men, the workplace might find itself more willing to make accommodations for the many workers who have family obligations and need flexibility to deal with sick children, or school schedules that do not conform to work schedules, as opposed to such accommodations being tied primarily just to mothers. Ideally, a push toward viewing parenting roles as a core feature in the lives of many men and women would help create a workplace culture that recognizes that, in addition to being stellar workers, their employees must also be able to fulfill their commitments as parents, a commitment that is shared across both mothers and fathers, across women and men.

Supplementary Material

Supplementary Material, park_online_appendix_1 – Leveraging the Social Role of Dad to Change Gender Stereotypes of Men

Supplementary Material, park_online_appendix_1 for Leveraging the Social Role of Dad to Change Gender Stereotypes of Men by Bernadette Park and Sarah Banchefsky in Personality and Social Psychology Bulletin

Supplementary Material

Supplementary Material, park_online_appendix_2 – Leveraging the Social Role of Dad to Change Gender Stereotypes of Men

Supplementary Material, park_online_appendix_2 for Leveraging the Social Role of Dad to Change Gender Stereotypes of Men by Bernadette Park and Sarah Banchefsky in Personality and Social Psychology Bulletin

Supplementary Material

Supplementary Material, park_online_appendix_3 – Leveraging the Social Role of Dad to Change Gender Stereotypes of Men

Supplementary Material, park_online_appendix_3 for Leveraging the Social Role of Dad to Change Gender Stereotypes of Men by Bernadette Park and Sarah Banchefsky in Personality and Social Psychology Bulletin

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by National Science Foundation Grant NSF1551099 awarded to Bernadette Park.

Supplemental Material

Supplementary material is available online with this article.

References

American National Election Studies. (2016). Retrieved from http://www.electionstudies.org/studypages/anes_pilot_2016/anes_pilot_2016_qnaire.pdf

Banchefsky

Park

(2016). The “new father”: Dynamic stereotypes of fathers. Psychology of Men and Masculinity, 17, 103-107.

Bosson

J. K.

Vandello

J. A.

Burnaford

R. M.

Weaver

J. R.

Arzu Wasti

(2009). Precarious manhood and displays of physical aggression. Personality and Social Psychology Bulletin, 35, 623-634.

Brandth

Kvande

(1998). Masculinity and child care: The reconstruction of fathering. The Sociological Review, 46, 293-313.

Chira

(2016, October 21). Men need help: Is Hillary Clinton the answer? The New York Times. Available from http://www.nytimes.com

Croft

Schmader

Block

(2015). An underexamined inequality: Cultural and psychological barriers to men’s engagement with communal roles. Personality and Social Psychology Review, 19, 343-370.

Crosby

(1984). The denial of personal discrimination. American Behavioral Scientist, 27, 371-386.

Diekman

A. B.

Eagly

A. H.

(2000). Stereotypes as dynamic constructs: Women and men of the past, present, and future. Personality and Social Psychology Bulletin, 26, 1171-1188.

Eagly

A. H.

Mladinic

(1989). Gender stereotypes and attitudes toward women and men. Personality and Social Psychology Bulletin, 15, 543-558.

10.

Eagly

A. H.

Mladinic

Otto

(1991). Are women evaluated more favorably than men? An analysis of attitudes, beliefs, and emotions. Psychology of Women Quarterly, 15, 203-216.

11.

Eagly

A. H.

Wood

(2012). Social role theory. In Van Lange

Kruglanski

Higgins

E. T.

(Eds.), Handbook of theories in social psychology (Vol. 2, pp. 458-476). doi:10.4135/9781446249222.n49

12.

Ganong

L. H.

Coleman

(1995). The content of mother stereotypes. Sex Roles, 32, 495-512.

13.

Ghavami

Peplau

L. A.

(2013). An intersectional analysis of gender and ethnic stereotypes: Testing three hypotheses. Psychology of Women Quarterly, 37, 113-127.

14.

Glick

Gangl

Gibb

Klumpner

Weinberg

(2007). Defensive reactions to masculinity threat: More negative affect toward effeminate (but not masculine) gay men. Sex Roles, 57, 55-59.

15.

Haines

E. L.

Deaux

Lofaro

(2016). The times they are a-changing . . . or are they not? A comparison of gender stereotypes, 1983–2014. Psychology of Women Quarterly, 40, 353-363. doi:10.1177/0361684316634081

16.

Hodges

A. J.

Park

(2013). Oppositional identities: Dissimilarities in how women and men experience parent versus professional roles. Journal of Personality and Social Psychology, 105, 193-216.

17.

Jost

J. T.

Glaser

Kruglanski

A. W.

Sulloway

F. J.

(2003). Political conservatism as motivated social cognition. Psychological Bulletin, 129, 339-375.

18.

Kitman

J. L.

(2016, December 19). Google wants driverless cars, but do we? The New York Times. Available from http://www.nytimes.com

19.

Koenig

A. M.

Eagly

A. H.

(2014). Evidence for the social role theory of stereotype content: Observations of groups’ roles shape stereotypes. Journal of Personality and Social Psychology, 107, 371-392.

20.

Mahalik

J. R.

Locke

B. D.

Ludlow

L. H.

Diemer

M. A.

Scott

R. P. J.

Gottfried

Freitas

(2003). Development of the Conformity to Masculine Norms Inventory. Psychology of Men and Masculinity, 4, 3-25.

21.

McPherson

Banchefsky

Park

(in press). Psychological consequences of the Dad Bod: Using biological and physical changes to increase essentialist perceptions of fathers. Journal of Experimental Social Psychology.

22.

Moss-Racusin

C. A.

(2014). Male backlash: Organizational penalties for men who violate gender stereotypes. In Burke

Major

(Eds.), Gender in organizations: Allies or adversaries to women’s career advancement (pp. 247-269). London, England: Edward Elgar.

23.

Park

Banchefsky

Reynolds

E. B.

(2015). Psychological essentialism, gender, and parenthood: Physical transformation leads to heightened essentialist conceptions. Journal of Personality and Social Psychology, 109, 949-967.

24.

Park

Smith

J. A.

Correll

(2010). The persistence of implicit behavioral associations for moms and dads. Journal of Experimental Social Psychology, 46, 809-815.

25.

Pleck

J. H.

(1997). Paternal involvement: Levels, sources, and consequences. In Lamb

M. E.

(Ed.), The role of the father in child development (3rd ed., pp. 66-103). New York, NY: John Wiley.

26.

Pleck

J. H.

Masciadrelli

B. P.

(2004). Paternal involvement by U.S. residential fathers: Levels, sources, and consequences. In Lamb

M. E.

(Ed.), The role of the father in child development (4th ed., pp. 222-271). Hoboken, NJ: John Wiley.

27.

Prentice

D. A.

Carranza

(2002). What women and men should be, shouldn’t be, are allowed to be, and don’t have to be: The contents of prescriptive gender stereotypes. Psychology of Women Quarterly, 26, 269-281.

28.

Raghunathan

T. E.

Rosenthal

Rubin

D. B.

(1996). Comparing correlated but nonoverlapping correlations. Psychological Methods, 1, 178-183. doi:10.1037/1082-989X.1.2.178

29.

Renfro

Duran

Stephan

W. G.

Clason

D. L.

(2006). The role of threat in attitudes toward affirmative action and its beneficiaries. Journal of Applied Social Psychology, 36, 41-74.

30.

Richard

Bond

Jr. Stokes-Zoota

(2003). One hundred years of social psychology quantitatively described. Review of General Psychology, 7, 331-363.

31.

Spears

Doosje

Ellemers

(1997). Self-stereotyping in the face of threats to group status and distinctiveness: The role of group identification. Personality and Social Psychology Bulletin, 23, 538-553.

32.

Stephan

W. G.

Renfro

C. L.

(2002). The role of threat in intergroup relations: The role of threats in intergroup relations. In Mackie

Smith

E. R.

(Eds.), From prejudice to intergroup emotions (pp. 191-208). New York, NY: Psychology Press.

33.

Troilo

Coleman

(2008). College student perceptions of the content of father stereotypes. Journal of Marriage and Family, 70, 218-227.

34.

Twenge

J. M.

(1997). Changes in masculine and feminine traits over time: A meta-analysis. Sex Roles, 36, 305-325.

35.

Vandello

J. A.

Bosson

J. K.

Cohen

Burnaford

R. M.

Weaver

J. R.

(2008). Precarious manhood. Journal of Personality and Social Psychology, 95, 1325-1339.

36.

Willer

Rogalin

C. L.

Conlon

Wojnowicz

M. T.

(2013). Overdoing gender: A test of the masculine overcompensation thesis. American Journal of Sociology, 118, 980-1022.

37.

Williams

J. E.

Best

D. L.

(1990). Measuring sex stereotypes: A multination study (Vol. 57). Newbury Park, CA: SAGE.

Supplementary Material

Please find the following supplemental material available below.

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

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

0.00 MB

0.18 MB

1.10 MB

2.15 MB