ArticlePDF Available

Evidence That Gendered Wording in Job Advertisements Exists and Sustains Gender Inequality

Authors:

Abstract and Figures

Social dominance theory (Sidanius & Pratto, 1999) contends that institutional-level mechanisms exist that reinforce and perpetuate existing group-based inequalities, but very few such mechanisms have been empirically demonstrated. We propose that gendered wording (i.e., masculine- and feminine-themed words, such as those associated with gender stereotypes) may be a heretofore unacknowledged, institutional-level mechanism of inequality maintenance. Employing both archival and experimental analyses, the present research demonstrates that gendered wording commonly employed in job recruitment materials can maintain gender inequality in traditionally male-dominated occupations. Studies 1 and 2 demonstrated the existence of subtle but systematic wording differences within a randomly sampled set of job advertisements. Results indicated that job advertisements for male-dominated areas employed greater masculine wording (i.e., words associated with male stereotypes, such as leader, competitive, dominant) than advertisements within female-dominated areas. No difference in the presence of feminine wording (i.e., words associated with female stereotypes, such as support, understand, interpersonal) emerged across male- and female-dominated areas. Next, the consequences of highly masculine wording were tested across 3 experimental studies. When job advertisements were constructed to include more masculine than feminine wording, participants perceived more men within these occupations (Study 3), and importantly, women found these jobs less appealing (Studies 4 and 5). Results confirmed that perceptions of belongingness (but not perceived skills) mediated the effect of gendered wording on job appeal (Study 5). The function of gendered wording in maintaining traditional gender divisions, implications for gender parity, and theoretical models of inequality are discussed.
Content may be subject to copyright.
Journal of Personality and Social Psychology
Evidence That Gendered Wording in Job Advertisements
Exists and Sustains Gender Inequality
Danielle Gaucher, Justin Friesen, and Aaron C. Kay
Online First Publication, March 7, 2011. doi: 10.1037/a0022530
CITATION
Gaucher, D., Friesen, J., & Kay, A. C. (2011, March 7). Evidence That Gendered Wording in
Job Advertisements Exists and Sustains Gender Inequality. Journal of Personality and Social
Psychology. Advance online publication. doi: 10.1037/a0022530
Evidence That Gendered Wording in Job Advertisements Exists and
Sustains Gender Inequality
Danielle Gaucher and Justin Friesen
University of Waterloo
Aaron C. Kay
Duke University
Social dominance theory (Sidanius & Pratto, 1999) contends that institutional-level mechanisms exist
that reinforce and perpetuate existing group-based inequalities, but very few such mechanisms have been
empirically demonstrated. We propose that gendered wording (i.e., masculine- and feminine-themed
words, such as those associated with gender stereotypes) may be a heretofore unacknowledged,
institutional-level mechanism of inequality maintenance. Employing both archival and experimental
analyses, the present research demonstrates that gendered wording commonly employed in job recruit-
ment materials can maintain gender inequality in traditionally male-dominated occupations. Studies 1 and
2 demonstrated the existence of subtle but systematic wording differences within a randomly sampled set
of job advertisements. Results indicated that job advertisements for male-dominated areas employed
greater masculine wording (i.e., words associated with male stereotypes, such as leader,competitive,
dominant) than advertisements within female-dominated areas. No difference in the presence of feminine
wording (i.e., words associated with female stereotypes, such as support,understand,interpersonal)
emerged across male- and female-dominated areas. Next, the consequences of highly masculine wording
were tested across 3 experimental studies. When job advertisements were constructed to include more
masculine than feminine wording, participants perceived more men within these occupations (Study 3),
and importantly, women found these jobs less appealing (Studies 4 and 5). Results confirmed that
perceptions of belongingness (but not perceived skills) mediated the effect of gendered wording on job
appeal (Study 5). The function of gendered wording in maintaining traditional gender divisions,
implications for gender parity, and theoretical models of inequality are discussed.
Keywords: inequality, intergroup relations, gender inequality, social dominance, belongingness
Despite widely touted egalitarian ideals, women in North Amer-
ica continue to be underrepresented in many areas of employment
including high levels of business, the natural sciences, and engi-
neering. In Canada, for example, less than 20% of engineering
undergraduates and only 9% of registered professional engineers
are women (Engineers Canada, 2010). A similar picture emerges
in the United States. Women comprise only 2.4% of Fortune 500
chief executive officers (Catalyst, 2008a), 20% of full professors
in the natural sciences (Catalyst, 2008b), and 11% of engineers
(U.S. Department of Labor, 2007). Why do women continue to be
underrepresented in these areas?
Individual-level factors that serve to keep women out of male-
dominated areas are well documented. Such factors manifest
within individuals in the form of beliefs, attitudes, and other
motivated tendencies. For example, system justification research
(see Jost & Banaji, 1994; Jost, Banaji, & Nosek, 2004) has dem-
onstrated that injunctification—people’s tendency to defend the
status quo via construing whatever currently is as natural and
desirable, and the way that things ought to be (Kay, Gaucher, et al.,
2009; Kay & Zanna, 2009)—is an individual-level process that can
account, at least in part, for women’s continued underrepresenta-
tion in male-dominated areas. Female participants who learned
about prevailing inequality (i.e., women’s underrepresentation in
the domains of business and politics) subsequently defended this
inequality as desirable and natural, an effect that was most pro-
nounced when system justification concerns were experimentally
heightened (Kay, Gaucher, et al., 2009).
Likewise, benevolent sexist beliefs (Glick & Fiske, 1996,
2001a, 2001b) and complementary (see Jost & Kay, 2005; Kay et
al., 2007) or compensatory (see Kay, Czaplin´ski, & Jost, 2009;
Kervyn, Yzerbyt, Judd, & Nunes, 2009; Napier, Thorisdottir, &
Jost, 2010) stereotypes are especially well suited to justify gender
inequalities. Endorsing the warm but incompetent stereotype of
housewives justifies women’s domestic role and exclusion from
the workplace (Cuddy, Fiske, & Glick, 2004; Fiske, Cuddy, Glick,
& Xu, 2002). Similarly, the competent but cold stereotype of
Danielle Gaucher and Justin Friesen, Department of Psychology, Uni-
versity of Waterloo, Waterloo, Ontario, Canada; Aaron C. Kay, Depart-
ment of Psychology and Neuroscience and Fuqua School of Business,
Duke University.
This research was prepared with the support of Social Sciences and
Humanities Research Council of Canada (SSHRC) Doctoral Fellowships to
Danielle Gaucher and Justin Friesen and research grants to Aaron C. Kay
from SSHRC and the Ontario Ministry for Innovation. We thank Fatima
Mitchell, Sandra Olheiser, and Gary Waller at Co-operative Education and
Career Services, University of Waterloo, for their valuable assistance with
Study 2.
Correspondence concerning this article should be addressed to Danielle
Gaucher, who is now at the Department of Psychology, Princeton Univer-
sity, Princeton, NJ 08540-1010, or Justin Friesen, Department of Psychol-
ogy, University of Waterloo, 200 University Avenue West, Waterloo,
Ontario N2L 3G1, Canada. E-mail: dgaucher@princeton.edu or
jp2fries@uwaterloo.ca
Journal of Personality and Social Psychology © 2011 American Psychological Association
2011, Vol. ●●, No. , 000– 000 0022-3514/11/$12.00 DOI: 10.1037/a0022530
1
working women has been used as justification for keeping women
out of (male-dominated) management positions (Fiske, Bersoff,
Borgida, Deaux, & Heilman, 1991; Phelan, Moss-Racusin, &
Rudman, 2008; Rudman, 1998; Rudman & Glick, 1999, 2001).
There is much less psychological research, however, document-
ing the institutional-level contributors to gender inequality.
Institutional-level contributors are those that manifest within the
social structure itself (e.g., public policy, law). According to social
dominance theory (SDT; Sidanius & Pratto, 1999), these
institutional-level mechanisms exist to reinforce and perpetuate
existing group-based inequality. Such contributors are often—
though certainly not always—so deeply embedded within the
social structure that they are overlooked by society at large
(Deutsch, 2006). These types of institutional-level factors remain
highly underresearched (Pratto, Sidanius, & Levin, 2006).
But despite the difficulty of detecting these systematic or insti-
tutional factors, their effects on individual-level psychological
processes are profound (e.g., increased antiegalitarianism, racism,
and victim blaming; Haley & Sidanius, 2005). Indeed, as Haley
and Sidanius (2005, p. 189) wrote:
Social hierarchies are in large part created, preserved, and recreated
by social institutions, or organizations. While lone individuals can
help to strengthen these hierarchies (e.g. by voting in favor of laws
that disproportionately handicap low-status groups) or to attenuate
them (e.g. by voting in favor of laws that instead help to level the
playing field), institutions should be able to impact hierarchies to a far
greater degree.
In the current research we identify an unacknowledged, institution-
level factor that may serve to reinforce women’s underrepresen-
tation in traditionally male-dominated occupations: gendered
wording used in job recruitment materials. Specifically, we inves-
tigate whether masculine-themed words (such as competitive,dom-
inate, and leader) emerge within job advertisements in male-
dominated areas, and whether the mere presence of these
masculine words dissuade women from applying to the area be-
cause they cue that women do not belong.
Job Advertisements as Institutional-Level
Contributors to Inequality
Women’s attrition in male-dominated fields, it has been pro-
posed, spikes at specific points along the career path, such as
between one’s master of science or master of arts degree and
doctorate, or at hiring and promotion (Holmes & O’Connell, 2007;
Tesch, Wood, Helwig, & Nattinger, 1995). In the geosciences, for
example, 38% of PhD graduates but only 26% of assistant profes-
sors are women (Holmes & O’Connell, 2007). It is plausible, then,
that institutional-level barriers to women’s participation in male-
dominated domains occur most prominently at certain critical
points. In the present research we focus on job recruitment as one
of those critical points.
Over 30 years ago, Bem and Bem (1973) investigated how job
advertisements that overtly specified a preference for male appli-
cants discouraged women from applying. They found that explicit
references to men as candidates for specific jobs and placing
advertisements in sex-segregated newspaper columns discouraged
men and women from applying to opposite-sex positions. In the
first of two seminal studies, participants were presented with a
series of job advertisements that were either sex biased (i.e., made
explicit reference to men as candidates for traditionally male-
dominated jobs such as lineman and women as candidates for
traditionally female jobs such as stewardess), unbiased (i.e., made
reference to both men and women as candidates), or sex reversed
(i.e., referred to women as ideal candidates for the typically male-
dominated jobs and men as ideal candidates for the traditionally
female jobs). The results were clear: Women were more interested
in male-dominated jobs when the advertisements were unbiased,
making reference to both men and women as candidates, than
when the advertisements made reference only to men (Bem &
Bem, 1973). Women reported the greatest interest in the male-
dominated jobs when the advertisements were sex reversed, ex-
plicitly referring to women as ideal candidates.
In a second study, female participants were presented with job
advertisements from a U.S. newspaper and asked to rate their
preference for each job. Half the participants read job advertise-
ments precisely as they appeared in the paper: sex segregated
under jobs–male and jobs–female columns. The other half read
identical advertisements, but this time they were integrated and
listed alphabetically with no sex labeling. Women preferred male-
dominated jobs when they were presented in the integrated rather
than the sex-segregated columns. Notably, this finding emerged
despite a disclaimer on both sets of advertisements citing that “job
seekers should assume that the advertiser will consider applicants
of either sex in compliance with the laws against discrimination”
(Bem & Bem, 1973, p. 15).
This type of bias in job advertisements, however, likely no
longer exists. On the heels of U.S. civil rights legislation (Title VII
of the 1964 Civil Rights Act) deeming this practice unconstitu-
tional, and the advent of the Equal Employment Opportunity
Commission, explicit sex segregation of advertisements had
abruptly ended by 1973 (Pedriana & Abraham, 2006). As a result,
it is no longer the case that job advertisements deter men or women
from applying to specific positions through explicit requests for
men or women or use of pronouns such as he or she. To many, this
suggested that this problem was solved.
However, although such explicit references to men or women as
ideal candidates have largely disappeared from the social land-
scape, it is possible that the gender of the ideal candidate is still
conveyed, but more subtly, through wording in the advertisement
that reflects broader cultural stereotypes about men and women. In
other words, even in the absence of explicit gender-biased direc-
tives, masculine and feminine themed words may be differentially
present in advertisements for jobs that are typically occupied by
males versus females, and the mere presence of this wording
difference may be sufficient to exert important downstream con-
sequences on individual-level appraisals of the relevant jobs.
The Nature of Subtle Wording Differences in
Job Advertisements
There is an established literature documenting widely held gen-
der stereotypes (e.g., Glick & Fiske, 1996) and differences in the
way men and women use everyday language (e.g., Pennebaker,
Mehl, & Niederhoffer, 2003). On the whole, women are perceived
as more communal and interpersonally oriented than men, whereas
men are more readily attributed with traits associated with leader-
ship and agency (Eagly & Karau, 1991; Heilman, 1983; Rudman
2GAUCHER, FRIESEN, AND KAY
& Kilianski, 2000). Moreover, gender differences in the linguistic
style of everyday speech are well documented (Carli, 1990; La-
koff, 1975). Women, for example, use a more communal style of
speech than men (Brownlow, Rosamond, & Parker, 2003; Haas,
1979; Leaper & Aryes, 2007) and make more references to social
and emotional words (Newman, Groom, Handelman, & Penne-
baker, 2008). Language use can also differ based on the gender of
whom one is writing about. An analysis of recommendation letters
for university faculty jobs within biology found that writers used
more “standout words” (e.g., outstanding,unique) when describ-
ing male than female candidates (Schmader, Whitehead, &
Wysocki, 2007). Similarly, Madera, Hebl, and Martin (2009) doc-
umented differential language use in recommendation letters for
university faculty jobs within psychology. Women were described
as more communal and less agentic than men, suggesting that
language use can unintentionally reflect stereotypical gender roles.
Furthermore, candidates whose letters contained more communal
traits were less likely to be hired, clearly demonstrating that these
gender-based differences in language use perpetuate inequality and
are not innocuous.
Drawing from these literatures, we reasoned that gendered
wording may emerge within job advertisements as a subtle mech-
anism of maintaining gender inequality by keeping women out of
male-dominated jobs. We predict that currently male-dominated
occupations will contain greater masculine wording in their job
advertisements than advertisements within female-dominated ar-
eas. For example, a job advertisement for a company in a male-
dominated area might, using masculine language, emphasize the
company’s “dominance” of the marketplace, whereas a company
in a less male-dominated area might, more neutrally, emphasize
the company’s “excellence” in the market. Likewise, a company
within a male-dominated occupation may be searching for some-
one to “analyze markets to determine appropriate selling prices,”
whereas an advertisement in a less male-dominated occupation
might emphasize “understanding markets to establish appropri-
ate selling prices” in its search. In both cases the job respon-
sibilities are similar, but the phrasing uses a more or less
masculine wording.
Origins of Gendered Wording Effects Within
Real-World Advertisements
Should we discover the hypothesized emergence of greater
masculine wording within advertisements for male-dominated
fields, two prominent social psychological theories, SDT (Sidanius
& Pratto, 2001) and social role theory (SRT; Eagly, 1987), suggest
the mechanisms for how this wording difference may have
emerged. However, although these two theories make some similar
predictions, they also diverge in important respects. SDT states
that “human societies tend to organize as group-based social
hierarchies” (Pratto et al., 2006, p. 272). One of the primary ways
societies produce and maintain group-based inequality, according
to the theory, is through institutional discrimination. From an SDT
perspective, then, gendered language used in job advertisements
likely serves as a covert institutional practice— one that is very
subtle—that ultimately serves to reinforce existing gender inequal-
ity, keeping women out of areas that men (the dominant group)
typically occupy.
SRT (Eagly, 1987) takes a different approach. Rather than focus
on gendered wording as an institutional-level mechanism keeping
women out of areas that men typically occupy, SRT posits that
gendered wording may arise from observations of differences in
role-based behavior. According to the theory, as women and men
engaged in traditional roles of homemaker and breadwinner, each
gender came to be associated with traits required of each role (i.e.,
nurturance and agency, respectively). Moreover, as a result of
these “original” gender roles, it is theorized that people enter
occupational areas typically associated with their traditional gen-
der role (e.g., women in nursing or men in firefighting). Thus,
according to SRT, the emergence of gendered words in job adver-
tisements is less the result of a motivated process in the service of
maintaining gender inequality, as SDT would predict, than the
result of an inference-based perceptual process whereby gendered
language emerges within advertisements depending on which gen-
der predominates. In other words, given that men are associated
with agency, if there are many men in a particular field, then traits
associated with men (i.e., agency) should emerge within the word-
ing of the advertisement. Likewise, if there are many women in a
particular field, then traits associated with women (i.e., commu-
nion) should be most likely to emerge within the wording of the
advertisement.
Both SRT and SDT, therefore, predict greater masculine word-
ing in male-dominated occupations than in female-dominated oc-
cupations, although for different underlying psychological reasons.
They differ, however, in their predictions for feminine wording
and female domains. Because SRT is an inference-based process,
it would predict the same type of effect for feminine wording as for
masculine wording: more feminine wording in female-dominated
jobs than in male-dominated jobs. SDT, in contrast, would not
necessarily predict a symmetric effect for masculine and feminine
wording, as the preservation of male dominance is much more
predicated on women being kept out of male domains than on men
being kept out of female domains. Across two naturalistic data sets
we content-coded job advertisements to empirically document
whether a novel institutional barrier to women’s inclusion in
traditionally male-dominated domains exists—that is, whether
gendered wording within real job advertisements emerges. In
addition, examining whether this effect operates symmetrically for
masculine and feminine wording across male- and female-
dominated domains may suggest which theory (SDT or SRT)
better accounts for the presence of gendered wording within real-
world job advertisements.
Crucially, we also predict that such differences in wording, if
they do in fact exist, may exert important effects on individual-
level judgments that facilitate the maintenance of inequality. Just
as other subtle variations in language can have a causal effect on
people’s behavior and attitudes (e.g., Boroditsky, 2001; Fitzsimons
& Kay, 2004; Hoffman & Tchir, 1990; Maass, 1999; Maass, Salvi,
Arcuri, & Semin, 1989; Newcombe & Arnkoff, 1979; Reitsma-van
Rooijen, Semin, & van Leeuwen, 2007), subtle variations in the
gendered wording used in advertisements may affect people’s
perception of jobs, such that men and women will find jobs
described in language consistent with their own gender most
appealing precisely because it signals they belong in that occupa-
tion. Specifically, we hypothesize that masculine wording likely
signals that there are many men in the field and alerts women to
the possibility that they do not belong.
3
GENDERED WORDING AND INEQUALITY
Gendered Wording as a Cue of Belongingness
There is ample evidence to suggest that belongingness—feeling
that one fits in with others within a particular domain—affects
people’s achievement motivation specifically and engagement
within a domain more generally, and that it can be signaled by cues
in the environment. Walton and Cohen (2007) provided some of
the most compelling evidence. They found that heightening
Blacks’ and Latinos’ sense of belongingness in academia by nor-
malizing their fears about not fitting in increased Blacks’ and
Latinos’ engagement with school, improving their grade point
averages and reducing dropout rates to levels comparable to that of
their White counterparts.
Other research has focused more specifically on the types of
cues that signal safety (i.e., belongingness) for members of under-
represented groups (e.g., Cheryan, Plaut, Davis, & Steele, 2009;
Davies, Spencer, Quinn, & Gerhardstein, 2002; Murphy, Steele, &
Gross, 2007; Purdie-Vaughns, Steele, Davies, Ditlmann, &
Crosby, 2008). Most relevant to the current research, Purdie-
Vaughns et al. (2008) demonstrated that cues that signal devalu-
ation of certain social identities can lead to domain disengagement.
In one study, reading a company newsletter that explicitly men-
tioned a “color-blind” philosophy but also showed photos suggest-
ing low minority representation led to less trust in and comfort
with the company, suggesting that belongingness can greatly affect
people’s tendency to approach particular domains.
Certainly, other factors also influence people’s propensity to
approach a domain. For example, practical considerations such as
having the skills required for the job and geographical location
factor into one’s decision to apply for a job. People are more likely
to apply to jobs for which they think they have the skills than for
ones for which they do not. But given the importance of belong-
ingness for domain engagement specifically (Walton & Cohen,
2007) and psychological well-being more generally (Baumeister &
Leary, 1995; Ryan & Deci, 2000), we expected anticipated feel-
ings of belongingness (as cued by gendered wording) to predict job
appeal independent of people’s perceptions of their own skill for
that job.
Overview of the Present Research
The present research has three main goals. First, via two large-
scale naturalistic studies in which we content-coded over 4,000 job
advertisements (Studies 1 and 2), we examine whether gendered
wording in job advertisements actually exists. This is the main
objective in Studies 1 and 2. Our main prediction is that we will
observe more masculine-themed words in male-dominated occu-
pations compared with female-dominated occupations, which
would be the type of institutional barrier that might prevent women
from entering traditionally male-dominated domains, as would be
predicted by SDT. In addition, by examining the usage of
feminine-themed words in male- and female-dominated occupa-
tions, we attempt to shed light on whether these wording differ-
ences do in fact reflect a means of social dominance or are best
explained by a cognitive inference approach, as would be sug-
gested by SRT.
Second, we examine the effects of gendered wording on indi-
viduals’ appraisals of occupations. In keeping with calls for a more
sociocultural approach to understanding inequality and oppression
(see Adams, Biernat, Branscombe, Crandall, & Wrightsman, 2008;
Deutsch, 2006), the current work directly tests how institutional-
level mechanisms— evident in gendered wording used in job ad-
vertisements—interact with individuals’ appraisals of the jobs and
perpetuate traditional divisions of labor. To do so, across a series
of experimental studies, we manipulate the masculine and femi-
nine wording of job advertisements and assess the causal effects of
gendered wording on perceptions of gender diversity, job appeal,
and anticipated feelings of belongingness. Should highly mascu-
line wording cue that women do not belong, and subsequently
make the job less appealing to women, then the current analysis
has important implications for the sex discrimination literature.
Third, we investigate the precise mechanism accounting for
gendered wording effects on job appeal. We hypothesize that
women would find jobs with masculinely worded advertisements
less appealing because masculine wording cues lower gender
diversity (Study 3) and signal to women that they do not belong in
such occupations (Studies 4 and 5), but not because they see
themselves as unskilled for the job (Study 5).
Study 1: Wording Differences in a Public Sample of
Online Job Advertisements
To investigate whether gendered wording differences in job
advertisements exist, we conducted two large naturalistic studies.
In Study 1, 493 randomly sampled online job advertisements from
typically male- and female-dominated occupations were coded for
the use of masculine and feminine words. In keeping with an SDT
perspective, we expected that advertisements within male-
dominated areas would contain a greater proportion of masculine
wording than advertisements within female-dominated areas. This
was our key hypothesis. However, by looking at the extent to
which this effect is mirrored for feminine wording within female
domains, we were also able to rule out an alternative interpretation:
that these effects are due to social role inferences rather than
mechanisms of social dominance. If social role processes are
underlying the emergence of gendered language in advertisements,
then we would also expect that advertisements within female-
dominated areas to contain more feminine words, compared with
male-dominated occupations, reflecting the greater number of
women in these occupations. The lack of such an effect, therefore,
would suggest that SRT processes are likely not at work.
1
Method
We obtained a list of occupations that included the proportion of
women and men in each job area (U.S. Department of Labor,
2007). Next, we identified Canada’s two leading job search web-
sites: monster.ca and workopolis.com. We then selected 11 occu-
pations that were highly male or female dominated and had a
corresponding category on both job websites. Based on these two
criteria, 11 occupations were selected for coding. Male-dominated
1
It is conceivable that SDT would also predict more feminine wording
in female-dominated areas, but this effect is not critical to an SDT expla-
nation. Without it, the pattern of results would still reflect an SDT effect,
because it is more important to keep subordinates out of dominants’ roles
than to keep dominants from freely choosing their roles. In contrast, SRT
clearly predicts the effect for feminine wording.
4GAUCHER, FRIESEN, AND KAY
jobs were plumber (1% women), electrician (2%), mechanic (2%),
engineer (11%), security guard (23%), and computer programmer
(26%); female-dominated jobs were administrative assistant
(97%), early childhood educator (94%), registered nurse (90%),
bookkeeper (90%), and human resources professional (71%). Us-
ing the job titles as search terms, we selected the first advertise-
ments listed for each occupation from both websites, to a maxi-
mum of 60 per occupation. This resulted in 493 advertisements
(231 representing male-dominated occupations and 262 represent-
ing female-dominated occupations; see Table 1). Managerial po-
sitions were not collected because of uncertainty about whether
managerial advertisements could be accurately categorized as male
or female dominated.
2
As a measure of gendered wording, lists of masculine and
feminine words were created with published lists of agentic and
communal words (e.g., individualistic,competitive,committed,
supportive; Bartz & Lydon, 2004; Rudman & Kilianski, 2000) and
masculine and feminine trait words (e.g., ambitious,assertive,
compassionate,understanding; Bem, 1974; Hoffman & Hurst,
1990; Schullo & Alperson, 1984; see Appendix A for a complete
list of the words that were coded). This is consistent with previous
research that has examined gender differences in language by
coding for specific words (Newman et al., 2008). Using content
analysis software (Pennebaker, Booth, & Francis, 2007), we gave
each advertisement masculine and feminine scores, representing
the percentage of total masculine and feminine words in each. A
score of 1.5% on masculine wording, for example, indicated that
1.5% of the total words in that advertisement were from the list of
masculine words. Masculine and feminine scores were correlated
but only weakly, r(493) .10, p.03.
Results and Discussion
Recall that we expected advertisements from male-dominated
occupations to contain greater masculine than feminine words,
compared with advertisements within female-dominated areas. To
test for this, we conducted a 2 (occupation: male dominated vs.
female dominated) 2 (wording: masculine vs. feminine) mixed
model analysis of variance (ANOVA), with wording as the re-
peated measure. A main effect of wording emerged, F(1, 491)
24.51, p.001,
p
2
.048, indicating that advertisements con-
tained more masculine words (M0.83%, SD 0.70%) than
feminine words (M0.63%, SD 0.75%). The main effect of
occupation was marginally significant, F(1, 491) 3.02, p.08,
p
2
.006, suggesting that there tended to be more coded words
within advertisements for male-dominated jobs (M0.77%,
SD 0.82%) than for female-dominated jobs (M0.69%, SD
0.65%).
Of greater interest, the predicted Wording Job Type interac-
tion emerged, F(1, 491) 18.33, p.001,
p
2
.036 (see
Figure 1). Masculine words were more likely to emerge within
advertisements for male-dominated jobs (M0.97%, SD
0.81%) than advertisements for female-dominated jobs (M
0.70%, SD 0.55%), t(491) 4.35, p.001, d0.40. There
was no difference in the presence of female words across male-
dominated (M0.57%, SD 0.77%) and female-dominated
occupations (M0.67%, SD 0.73%), t(491) 1.48, p.14,
d0.13.
The above analysis confirmed the most straightforward predic-
tion: Job advertisements within male-dominated areas contained
greater masculine wording than advertisements from female-
dominated areas. A more nuanced test of the hypothesis, however,
is to examine whether differences in wording also vary continu-
ously as a function of the number of men present in the occupation.
Are greater numbers of men in any given occupation associated
with more masculine and fewer feminine words? Indeed, this was
the case. Across all advertisements, the percentage of men in an
occupation (according to the U.S. Department of Labor, 2007) was
used to predict the percentages of masculine and feminine words
emerging across all the advertisements. More men in an occupa-
tion was associated with a greater presence of masculine words,
␤⫽.17, t(492) 3.74, p.05, and a marginally, though not
significant, lower presence of feminine words, ␤⫽⫺.19, t(492)
1.90, p.06.
3
Study 2: Wording Differences in a University Sample
of Co-Op Job Advertisements
Study 1 provided initial evidence that gendered wording in job
advertisements exists. Consistent with both SDT and SRT, adver-
tisements from male-dominated fields contained greater masculine
wording than advertisements from female-dominated fields. More
in keeping with an SDT perspective, however, there was no
difference in the presence of feminine wording across male- and
female-dominated fields, as an SRT interpretation would suggest
that there should have been. In Studies 3–5, we experimentally
assess the social psychological effects of this type of wording bias.
Before doing so, however, we sought to replicate the findings
using a different sample to ensure that the wording differences that
emerged were not based on something idiosyncratic about the
particular 11 occupations selected in Study 1. To do this, we
2
Due to differences in the number of advertisements posted online,
the number of advertisements collected for each occupation varied from
14 to 59.
3
We regressed a ranked variable corresponding to the exact percentage
of men within each occupation (as defined by the U.S. Department of
Labor, 2007) on the masculine wording and feminine wording variables.
Table 1
Composition of Coded Advertisements, Study 1
Advertisements
Occupation N%
Male dominated 231 47
Plumber 36 7
Electrician 55 11
Mechanic 14 3
Engineer 59 12
Security guard 22 4
Computer programmer 45 9
Female dominated 262 53
Human resources professional 45 9
Bookkeeper 55 11
Registered nurse 57 11
Early childhood educator 50 10
Administrative assistant 55 11
5
GENDERED WORDING AND INEQUALITY
randomly sampled job advertisements for university students that
were targeted at stereotypically male- and female-dominated fac-
ulties (e.g., engineering vs. arts, respectively) but that were from
hundreds of employers. The University of Waterloo has the largest
cooperative education program in Canada. Students in the co-op
program alternate between terms of on-campus coursework and
off-campus work placements. Each term hundreds of companies
advertise thousands of student jobs that students can access
through a university website. We were granted access to the text
of 3,640 of these job advertisements targeted at six university
faculties.
In contrast to the online advertisements in Study 1, which were
organized by occupation title (e.g., engineer, plumber), the co-op
job advertisements were organized by university faculty (e.g.,
engineering, arts, applied health sciences). Employers could
choose to target their advertisements to students in engineering,
arts, or another faculty. There is generally a correspondence be-
tween the faculty targeted and the domain of the job; for example,
jobs targeted at students in the engineering faculty are almost
always within the engineering field, but the job titles themselves
within that faculty varied. By categorizing advertisements by fac-
ulty instead of by specific job titles, it ensured that the wording
effects were not particular to jobs titled “engineer” or “computer
programmer” but reflected a general trend within the engineering
or computer science fields. Again, in keeping with SDT and SRT,
we expected to find more masculine wording in advertisements
targeted at students in male-dominated faculties (i.e., from more
male-dominated fields) than in advertisements targeted at students
in female-dominated faculties. Furthermore, on the basis of the
findings from Study 1, we expected to find no difference in the
amount of feminine wording found across subject areas.
Method
Job advertisements (N3,640) were randomly selected from
the on-campus co-operative job posting site at the University of
Waterloo. Given the university’s emphasis on engineering, math,
and computer science programs, there were more job advertise-
ments targeted to typically male-dominated faculties (n3,116)
than to typically female-dominated faculties (n524). The job
advertisements targeted to typically male-dominated faculties were
sampled from engineering (n1,682), math and computer science
(n920), science (n205), business and economics (n209),
and accounting and financial management (n100). The job
advertisements targeted to typically female-dominated faculties
were from applied health studies (n205), arts (n160), and
environmental studies (n159). All the faculties that were
represented on the co-op job posting site were included in the
present analyses. As in Study 1, we coded each advertisement for
the percentage of masculine and feminine words contained in the
advertisement using linguistic software (Pennebaker et al., 2007).
In this sample, masculine and feminine scores were again corre-
lated but only weakly, r(3640) .05, p.001.
Results
To test for wording differences, we conducted a 2 (faculty
advertised: male dominated vs. female dominated) 2 (wording:
masculine vs. feminine wording) mixed model ANOVA, with
wording as the repeated measure. A main effect of wording
emerged, F(1, 3638) 191.67, p.001,
p
2
.050, indicating
that advertisements, overall, contained more masculine words
(M1.09%, SD 0.88%) than feminine words (M0.67%,
SD 0.70%). A main effect of faculty also emerged, F(1, 3638)
18.76, p.001,
p
2
.005, indicating there were more coded
words within advertisements targeted at male-dominated faculties
(M0.90%, SD 0.83%) than within advertisements targeted at
female-dominated faculties (M0.78%, SD 0.73%).
Of greater interest, once again the predicted Wording Job
Type interaction emerged, F(1, 3638) 14.76, p.001,
p
2
.004 (see Figure 2). As expected, there was a greater percentage of
masculine words within advertisements for male-dominated facul-
ties (M1.12%, SD 0.89%) than within advertisements for
female-dominated faculties (M0.91%, SD 0.76%), t(3638)
5.23, p.001, d0.24. There was no difference in the presence
of feminine words between male-dominated (M0.67%, SD
0.70%) and female-dominated faculties (M0.65%, SD
0.68%), t(3638) 0.56, p.56, d0.03.
Discussion: Studies 1 and 2
Across two independent samples, job advertisements for male-
dominated occupations contained more stereotypically masculine
Figure 1. Mean percentage of gendered wording as a function of occu-
pation area (Study 1). Error bars indicate standard error.
Figure 2. Mean percentage of gendered wording as a function of occu-
pation area (Study 2). Error bars indicate standard error.
6GAUCHER, FRIESEN, AND KAY
words than job advertisements for female-dominated occupations.
Furthermore, in both cases there was no difference in the presence
of stereotypically feminine words across male- and female-
dominated occupations. This lack of an effect for feminine word-
ing suggests that gendered wording, at least in the current studies,
is not simply the result of a perceptual process whereby people
make inferences about the traits required for jobs based on roles
traditionally held by men and women. Next, we sought to test
whether these types of wording differences affect individual-level
psychological functioning in a way that can lead to the perpetua-
tion of gender divisions. Does the inclusion of more masculine
stereotype-linked words, like the ones we systematically found in
the advertisements for male-dominated jobs, make a given job less
appealing to individual female candidates? Three laboratory ex-
periments (Studies 3–5) tested the causal effects of gendered
wording on perceptions of gender diversity, anticipated feelings of
belongingness, and, ultimately, job appeal.
Study 3: The Effect of Gendered Wording on
Diversity Perceptions
In Study 3 participants read a series of job advertisements
manipulated to be masculinely or femininely worded and then
estimated the perceived gender diversity within the various occu-
pations. We created advertisements for traditionally male-
dominated (engineer, plumber) and female-dominated (registered
nurse, administrative assistant) occupations, as well as two neutral
jobs (real estate agent, retail sales manager). The neutral jobs
allowed for testing of gendered wording effects within occupations
not as strongly associated with either gender stereotype. Unsur-
prisingly, regardless of an advertisement’s wording, people will
likely perceive more men within some occupations (e.g., engineer-
ing) than other occupations (e.g., nursing). We hypothesized, how-
ever, that gendered wording will have an additional effect, inde-
pendent from the type of job, such that people would perceive
more men within jobs that were masculinely worded.
Method
Forty-three Canadian-born introductory psychology students
participated online (28 women, 15 men; 65% White, 21% Asian,
5% Indian, and 9% other or not listed) in exchange for course
credit.
Using a within-subjects design, we provided each participant
with six job advertisements to read. Two were from each job type
(male dominated, female dominated, neutral). Within each job
type, one advertisement used more feminine wording, and the
other used more masculine wording. To create the advertisements,
we used common phrases from the advertisements sampled in
Study 1 to create a generic advertisement for each job (70 –115
words). Each included a company description, job description, and
qualifications. Wording was manipulated by selectively substitut-
ing masculine and feminine words from the lists used in Studies 1
and 2. This resulted in two versions of each advertisement, one
masculinely worded and one femininely worded.
4
For example,
the masculinely worded advertisement for a registered nurse
stated, “We are determined to deliver superior medical treatment
tailored to each individual patient,” whereas the femininely
worded advertisement stated, “We are committed to providing top
quality health care that is sympathetic to the needs of our patients”
(emphasis added). Advertisements are included in Appendix B.
To control for any idiosyncrasies of specific jobs (e.g., a general
preference for engineering over plumbing) that might be con-
founded with wording, we counterbalanced across individual oc-
cupations. For half the participants, the masculinely worded ad-
vertisements were plumber, retail sales manager, and registered
nurse (with the three remaining advertisements femininely
worded). For the other half, the masculinely worded advertise-
ments were engineer, real estate agent, and administrative assistant
(with the three remaining advertisements femininely worded).
Presentation order was also counterbalanced.
After each advertisement, participants completed two items
assessing their perception of the number of women within each
job that they read. Specifically, participants were asked, “How
many women . . .” (a) “work in this company?” and (b) “work
in the position being advertised?” on a Likert scale ranging
from 0 (0% women)to20(100% women), each point labeled in
5% increments.
Furthermore, after each advertisement participants were asked
which factors had affected their perceptions of that advertisement;
not one participant mentioned wording.
Results and Discussion
To test whether masculine and feminine wording actually af-
fects participants’ perceptions of the gender diversity within each
occupation, we conducted a 3 (job type: male dominated vs.
neutral vs. female dominated) 2 (wording: masculine vs. fem-
inine) 2 (participant gender: male vs. female) mixed model
ANOVA, with job type and wording as repeated measures and
gender between-subjects. First, we report the effects not associated
with the wording predictions. For ease of interpretation, raw scale
scores are also reported as percentages of women in square brack-
ets. A main effect of job type emerged, F(2, 80) 199.32, p
.001,
p
2
.833, indicating, as expected, that traditionally male-
dominated occupations were estimated to have fewer women (M
4.27 [20%], SD 3.31 [16%]) than neutral (M12.35 [59%],
SD 3.03 [14%]), t(80) 16.12, p.001, d2.62, and
female-dominated occupations (M12.35 [59%], SD 3.49
[17%]), t(80) 16.50, p.001, d2.13. Estimates of neutral
and female-dominated occupations did not differ from one another,
t(80) 1.19, p.24, d0.00. This main effect, however, was
qualified by a significant Job Type Gender interaction, F(2,
80) 5.00, p.05,
p
2
.011, reflecting the tendency for female
participants (M13.20 [63%], SD 2.83 [13%]) to estimate
greater numbers of women in neutral occupations than estimated
by male participants (M10.64 [51%], SD 2.73 [13%]; see
Table 2).
Of most relevance, the predicted wording main effect also
emerged, F(1, 40) 10.40, p.02,
p
2
.206. People perceived
4
As a manipulation check, each advertisement was coded with the same
procedure as in Studies 1 and 2, confirming that the six masculinely
worded advertisements had more masculine (M8.40%, SD 1.75%)
than feminine words (M0.00%, SD 0.00%), t(5) 12.41, p.05,
and the six femininely worded advertisements had more feminine (M
6.86%, SD 1.67%) than masculine words (M0.15%, SD 0.36%),
t(5) 8.45, p.05.
7
GENDERED WORDING AND INEQUALITY
fewer women within the occupations advertised with masculine
wording (M9.06 [43%], SD 4.80 [23%]) than occupations
advertised with feminine wording (M10.26 [49%], SD 5.20
[25%]). Wording did not interact with gender or job type ( ps
.10), which suggests that participants perceived more men within
jobs that had masculinely worded advertisements than jobs that
had femininely worded advertisements, regardless of participant
gender or whether that occupation was traditionally male or female
dominated.
Study 4: Job Appeal and Belongingness
In Studies 1 and 2 we found that male-dominated jobs tend to
employ more masculine wording in their recruitment materials.
Study 3 demonstrated that these wording differences may not be
entirely innocuous: Masculine wording led people to predict that
there are relatively more men within the relevant occupation. In
Study 4 we investigated another consequence of gendered word-
ing, one that is more crucial in order for these institutional-level
biases to actually maintain gender inequality: that of job appeal.
We hypothesized that masculine wording may actually reduce
women’s interest in a job because it signals to them that they may
not belong. To test this hypothesis, we used the same experimental
materials as in Study 3. This time, however, rather than assess
people’s perception of the number of women in a given occupa-
tion, we assessed job appeal and anticipated belongingness.
Method
One hundred and two English-fluent introductory psychology
students participated online for course credit. Six were excluded
(four failed to specify their gender, and two were outliers), leaving
96 participants (63 women, 33 men; 45% White, 31% Asian, 10%
Indian, and 13% other or not listed).
Using the same job advertisements and within-subjects design
as in Study 3, we gave each participant six job advertisements to
read. As in Study 3, presentation order was counterbalanced. There
was no effect of order.
After each advertisement participants completed six items as-
sessing job appeal (“This job is appealing”; “I think I could enjoy
this job”; “This is not a job I would want,” reverse coded; “This
company would be a good employer”; “This job looks interesting”;
and “This company seems like a great place to work”; alphas
ranged from .90 to .94 for each advertisement). Four items assess-
ing anticipated belongingness (“I could fit in well at this com-
pany”; “I’m similar to the people who work in this career”; “My
values and this company’s values are similar”; and “The type of
people who would apply for this job are very different from me,”
reverse coded). The four-item measure of belongingness was cre-
ated by adapting items from Walton and Cohen’s (2007) Belong-
ingness Scale, making them specific to the occupational domain.
Items 1–3 were virtually identical to items used by Walton and
Cohen (2007). Item 4 was designed to more fully assess the
similarity component of belongingness. All four items showed
good reliabilities, with alphas ranging from .80 to .86 for each
advertisement. All items used a Likert scale ranging from 1
(strongly disagree)to7(strongly agree).
Last, after reading all the advertisements, participants ranked
them from most appealing to least appealing. Finally, as in Study
3, after each advertisement participants were asked which factors
had affected their perceptions of that advertisement; not one par-
ticipant mentioned wording.
We predicted a Wording Gender interaction, in which women
judge masculinely worded jobs less appealing than femininely
worded jobs. For men, there are two possibilities. On the one hand,
they might show the opposite pattern and prefer masculinely
worded jobs to femininely worded jobs. On the other hand, higher
status groups—such as men—might be less concerned with cues of
belongingness because of their already secure high status in the
gender hierarchy. Thus, gendered wording may not affect men’s
job appeal ratings. We also tested whether gendered wording
effects on job appeal are driven, at least in part, by expected
belongingness.
Results
Subsequent analyses used a 3 (job type: male dominated vs.
neutral vs. female dominated) 2 (wording: masculine vs. fem-
inine) 2 (participant gender: male vs. female) mixed measures
ANOVA, with job type and wording as repeated measures and
gender between-subjects.
Job appeal: Likert ratings. First we report analyses that are
not central to our main hypotheses. A main effect of job type
emerged, F(2, 188) 4.57, p.01,
p
2
.046. In general,
male-dominated occupations were less appealing (M4.07, SD
1.45) than either neutral (M4.50, SD 1.30), t(188) 2.57,
p.02, d0.33, or female-dominated occupations (M4.50,
SD 1.39), t(188) 2.71, p.009, d0.34. Appeal ratings of
the neutral and female-dominated occupations did not differ from
one another (t1). No other main effects emerged.
Of primary importance, the predicted Wording Gender inter-
action emerged, F(1, 94) 9.70, p.002,
p
2
.094 (see
Figure 3). Women found occupations that were masculinely
worded less appealing (M4.16, SD 1.41) than femininely
worded occupations (M4.50, SD 1.42), F(1, 62) 6.74, p
.01,
p
2
.098. Men only marginally showed the opposite pattern,
tending to rate masculinely worded advertisements (M4.61,
SD 1.34) as more appealing than femininely worded advertise-
ments (M4.22, SD 1.29), F(1, 32) 3.58, p.07,
p
2
.100. Did the effect of wording vary between job types (tradition-
ally male vs. female)? The absence of a three-way interaction (F
Table 2
Perceived Gender Diversity Within Occupations as a Function
of Participant Sex, Study 3
Gender
Male-dominated
occupations
Neutral
occupations
Female-dominated
occupations
Female
M4.46
a
(21%) 13.20
b
(63%) 12.26
b
(58%)
SD 3.50 (17%) 2.83 (13%) 3.47 (17%)
Male
M3.93
a
(19%) 10.64
b
(51%) 12.54
c
(60%)
SD 3.05 (15%) 2.73 (13%) 3.59 (17%)
Note. Means and standard deviations reflect the 21-point scale seen by
participants. For ease of interpretation, raw scale scores are also reported in
parentheses as percentages of women. Within rows, different subscripts
indicate means differ ( p.05) by least significant difference test.
8GAUCHER, FRIESEN, AND KAY
1) suggests no; regardless of whether the occupations were tradi-
tionally more male or female dominated, participants found jobs
most attractive when there was a match between their gender and
the gendered wording used in the advertisement.
Job appeal: Relative rankings. Rankings were coded so that
higher rankings indicated more appeal (e.g., 6 indicated the most
appealing job, 1 indicated the least appealing).
5
A main effect of
job type emerged, F(2, 180) 9.45, p.05,
p
2
.095.
Participants ranked male-dominated occupations as less appealing
(M2.88, SD 1.62) than neutral (M3.80, SD 1.59),
t(180) 4.00, p.001, d0.61, and female-dominated occu-
pations (M3.82, SD 1.75), t(180) 4.17, p.001, d0.53.
This effect was qualified by a significant Job Type Gender
interaction, F(2, 180) 7.21, p.001,
p
2
.074, indicating that
women ranked male-dominated jobs (M2.58, SD 1.49) lower
than neutral (M3.97, SD 1.57), t(118) 6.40, p.001, d
0.86, and female-dominated jobs (M3.95, SD 1.70), t(118)
6.73, p.001, d0.82, whereas appeal rankings of the neutral
and female-dominated occupations did not differ from one another
(t1). Men ranked all three job types equally (M
male
3.44,
SD 1.73; M
neutral
3.50, SD 1.61; M
female
3.56, SD
1.82; all ts1).
Importantly, the predicted Wording Gender interaction again
emerged, F(1, 90) 8.68, p.004,
p
2
.088. Women ranked
masculinely worded jobs less favorably (M3.25, SD 1.74)
than femininely worded jobs (M3.75, SD 1.64), F(1, 59)
7.74, p.007,
p
2
.120. For men there was a nonsignificant
opposite trend, to prefer masculinely worded jobs (M3.75,
SD 1.78) over femininely worded jobs (M3.25, SD 1.61),
F(1, 31) 2.43, p.13,
p
2
.073.
Belongingness. First we report analyses not central to our
main hypotheses. A main effect of job type emerged, F(2, 188)
4.84, p.009,
p
2
.049, so that, overall, less anticipated
belongingness was reported in neutral and male-dominated jobs
than in female-dominated jobs, but this was qualified by a mar-
ginally significant Job Type Gender interaction, F(2, 188)
2.86, p.06,
p
2
.029, so that only women anticipated less
belongingness within male-dominated occupations, whereas men
showed no differences. Means and follow-up tests are reported in
Table 3.
Most relevant to the main hypothesis, the predicted Wording
Gender interaction emerged, F(1, 94) 6.16, p.02,
p
2
.061.
Women reported greater anticipated belongingness within occupa-
tions that were femininely than masculinely worded (M4.31,
SD 1.35 vs. M3.98, SD 1.29, respectively), F(1, 62)
6.35, p.01,
p
2
.093. For men, there was no effect of gendered
wording, with no difference between femininely and masculinely
worded occupations (M3.99, SD 1.16 vs. M4.23, SD
1.31, respectively; p.22).
Women found femininely worded advertisements more appeal-
ing than masculinely worded advertisements. Next we tested
whether this difference in belongingness drove the observed dif-
ferences in appeal. Because this is a within-subjects design, dif-
ference scores were calculated for job appeal (Likert ratings) and
anticipated belongingness by subtracting the mean rating for fem-
ininely worded advertisements from the mean rating for mascu-
linely worded advertisements (following Judd, Kenny, & Mc-
Clelland, 2001). Therefore, for both difference scores a higher
value indicated a preference (i.e., more appeal or more belonging-
ness) for masculinely worded jobs over femininely worded jobs.
We tested mediation using structural equation modeling (AMOS
16.0) and the procedures outlined in Sadler, Ethier, and Woody (in
press); MacKinnon, Lockwood, and Williams (2004); and Shrout
and Bolger (2002). Using 5,000 bootstrapped samples, we calcu-
lated path estimates and bias-corrected 95% confidence intervals
(CIs) of the indirect effect of gender on job appeal via belonging-
ness. Gender (dummy coded: female 1, male 0) predicted the
anticipated belongingness difference score (␤⫽⫺.25, p.01), so
that women anticipated less belongingness within the jobs adver-
tised with masculine wording. This reduction in perceived belong-
ingness predicted less job appeal (␤⫽.72, p.001) and was a
significant indirect effect (␤⫽⫺.18, 95% CI [.04, .32], p
.02). Because the 95% CI of the indirect effect did not include
zero, we can be confident (␣⫽.05) that the women found
masculinely worded advertisements less appealing, at least in part,
because of the corresponding decrease in anticipated belonging-
ness.
Discussion
Beyond affecting people’s perception of gender diversity within
an occupation (Study 3), increases in masculine wording were
5
For the relative rankings, four additional participants were excluded
because they ranked the same job multiple times (e.g., as both first and
third).
Figure 3. Job appeal as a function of gendered wording and participant
gender (Study 4), on a scale of 1 to 7. Error bars represent standard error.
Table 3
Belongingness as a Function of Participant Sex and Job Type,
Study 4
Male-
dominated
occupations
Neutral
occupations
Female-
dominated
occupations
Gender M SD M SD M SD
Female 3.79
a
1.31 4.24
b
1.28 4.41
b
1.33
Male 4.10
b
1.24 3.95
b
1.15 4.28
b
1.32
Note. Within rows, different subscripts indicate means differ ( p.05)
by least significant difference test, on a scale of 1 to 7.
9
GENDERED WORDING AND INEQUALITY
sufficient to decrease women’s job appeal ratings and their antic-
ipated belongingness in specific occupations. Men displayed only
a slight preference for masculinely worded advertisements, and
gendered wording did not affect men’s anticipated feelings of
belongingness. This latter finding supports our contention that
masculine wording reflected in real job advertisements primarily
serves to keep women out of the areas that men typically occupy.
In contrast, women were deterred from masculinely worded jobs,
finding them less appealing, compared with the same types of jobs
advertised with feminine wording. These effects occurred both
when the jobs were evaluated independently (in the case of the job
appeal Likert measure) and, more strongly, when participants
explicitly compared them with one another (the job appeal ranking
measure). Strikingly, the type of job that women read about (i.e.,
engineer, real estate broker, nurse) did not interact with wording.
Across all types of jobs, women ranked jobs more negatively when
they were written with masculine wording.
In addition, women anticipated less belongingness in jobs that
were masculinely worded, and these changes in anticipated be-
longingness mediated the effects on appeal. Thus, this study also
provides evidence for the hypothesis that masculine wording is
unappealing to women precisely because it conveys that they may
not belong in that job. In Study 5 we sought further evidence for
this hypothesis and sought to rule out a potential alternative driver
of the observed effects.
Study 5: The Effect of Gendered Wording on Job
Appeal, Personal Skill, and Belongingness
Studies 3 and 4 demonstrated the effects of gendered wording
on job appeal and suggested that perceptions of gender diversity
and anticipated belongingness may be driving those effects. How-
ever, despite the general importance of belongingness (Baumeister
& Leary, 1995; Ryan & Deci, 2000; Walton & Cohen, 2007), it
may not be the sole driver of gendered wording effects. One could
argue that the manipulation of gendered wording changed the
qualifications that people think are necessary for the job and that
these differences in qualifications and skills, in turn, affected the
appeal of the various occupations. Masculine wording, for exam-
ple, may cue that different skills or training are required for the
job, and this assessment subsequently makes the job less appeal-
ing. This alternative account would suggest that the effects of
gendered wording on job appeal are not necessarily driven by
psychological variables (such as belongingness) that are ultimately
irrelevant to a job’s objective tasks, as we have proposed here, but
are instead due to logical assessments of whether one’s personal
skills match the job’s requirements. Thus, in Study 5 items assess-
ing whether people thought that they have the personal ability and
skills to perform each job (assuming they met the education and
experience requirements) were added, ensuring that anticipated
belongingness remains a significant contributor even when con-
trolling for perceived skill.
In Study 5 we assessed the effects of gendered wording on
gender diversity, belongingness, personal skills, and job appeal
using a between-subjects design, rather than the within-subjects
design used in Studies 3 and 4, and within a domain that does not
have strong preexisting norms about which gender has the best
skills for the job. Given our primary interest in how gendered
wording affects women’s perceptions of jobs and our focus on
whether gendered wording can affect women’s appraisals of jobs
in areas without strong preexisting gender stereotypes, we re-
cruited all female participants in the current study. Here female
participants read just one of two advertisements, for a real estate
agent, that was either masculinely or femininely worded. Next,
they responded to items assessing gender diversity, belongingness,
personal skill, and job appeal. The real estate agent advertisement
was an ideal context to test whether the effects of gendered
wording are strong enough to women’s appeal ratings of jobs even
in the absence of strong predetermined norms about who best
belongs or has the skills necessary for that job. We expected
women to rate the real estate job as less appealing when it was
masculinely than femininely worded, and for this effect to be
mediated by anticipated belongingness.
Method
One hundred and eighteen Canadian-born introductory psychol-
ogy students participated online in exchange for course credit.
Participants were all female (65% White, 15% Asian, 8% Indian,
and 12% other or not listed).
Each participant read one of two real estate agent advertise-
ments previously used in Studies 3 and 4. One version of the
advertisement used more feminine wording, and the other used
more masculine wording (see Appendix B). Participants then com-
pleted our dependent measures. First was the index of gender
diversity, which was the same two items used in Study 3: “How
many women . . .” (a) “work in this company?” and (b) “work in
the position being advertised?” on a Likert scale ranging from 0
(0% women)to10(100% women), each point labeled in 10%
increments, r(119) .44, p.001. Next, participants were asked
three items assessing their personal skill for the job (“I could
perform well at this job”; “If I had this job I would definitely
succeed at it”; and “If I had this job there is a good chance I would
fail at it,” reverse coded; ␣⫽.81), belongingness (e.g., “I could fit
in well at this company”; ␣⫽.76), and job appeal (e.g., “This job
is appealing”; ␣⫽.85) on a Likert scale ranging from 1 (strongly
disagree)to7(strongly agree). The anticipated belongingness and
job appeal items were identical to those used in Study 4, and
moderately correlated, r(119) .49, p.001.
Results and Discussion
All the subsequent analyses used ANOVA with wording (fem-
inine vs. masculine) as a between-subjects factor. As expected,
women perceived fewer women within the masculinely than fem-
ininely worded real estate advertisements (M4.27 [39%], SD
1.28 [12%] vs. M5.27 [48%], SD 1.31 [12%], respectively),
F(1, 117) 17.83, p.001,
p
2
.132. Critically, women rated
the real estate agent job as less appealing when it was masculinely
than femininely worded (M3.86, SD 1.22 vs. M4.34,
SD 1.07, respectively), F(1, 117) 5.15, p.03,
p
2
.042,
and reported less belongingness when it was masculinely than
femininely worded (M3.71, SD 1.04 vs. M4.09, SD
0.89, respectively), F(1, 117) 4.42, p.04,
p
2
.036. Wording
of the advertisement did not affect women’s perceived level of
personal skill required for the job (M
masculine
4.89, SD 1.29;
M
feminine
4.99, SD 1.09; F1).
10 GAUCHER, FRIESEN, AND KAY
To test whether feelings of anticipated belongingness mediated
the effect of gendered wording on job appeal, we again used
structural equation modeling (AMOS 16.0) and the procedures
outlined in Sadler et al. (in press), MacKinnon et al. (2004), and
Shrout and Bolger (2002). We tested the model shown in Figure 4:
the effect of wording on job appeal using anticipated belonging-
ness and perceived skill as simultaneous mediators (5,000 boot-
strapped samples). Controlling for perceived skill, advertisement
wording (dummy coded: feminine wording 0, masculine word-
ing 1) predicted anticipated belongingness (␤⫽⫺.19, p.04),
so that women anticipated less belongingness within the jobs
advertised with masculine wording. This reduction in perceived
belongingness predicted less job appeal (␤⫽.67, p.001). With
bias-corrected 95% CIs, the indirect effect of advertisement word-
ing on appeal via belongingness was significant (␤⫽⫺.13, 95%
CI [.01, .24], p.03). Again, women found masculinely
worded advertisements less appealing, at least in part, because of
the corresponding decrease in anticipated belongingness.
The perceived skill measure was associated with job appeal,
suggesting that it was an adequately reliable measure, but its
mediating effect was not significant. That is, women who reported
that they had the personal skill needed for a job did find it more
appealing, but their ratings of personal skill were not affected by
gendered wording. These results are consistent with the hypothesis
that gendered wording signals who belongs and who does not, and
this, in part, affects the appeal of a job, independent of whether one
perceives one has the personal skill to perform that job.
6
General Discussion
The results of these studies demonstrate that masculine wording
in job advertisements leads to less anticipated belongingness and
job interest among women, which, we propose, likely perpetuates
gender inequality in male-dominated fields. Study 1, a large-scale
naturalistic study, demonstrated that advertisements for male-
dominated jobs contained greater masculine wording than adver-
tisements for female-dominated jobs. Study 2 replicated this word-
ing effect in a different context: Advertisements posted within
faculties associated with male-dominated fields contained more
masculine wording than advertisements posted within faculties
associated with female-dominated fields. No difference in the
presence of feminine wording between male- and female-
dominated occupations emerge in either Study 1 or Study 2, and
we hypothesize such differences should have been found if social
role processes were driving the emergence of gendered wording.
Rather, the pattern of results across Studies 1 and 2 is more
supportive of a social dominance perspective, which predicts an
asymmetric pattern of wording differences.
Moreover, at the individual level, masculine wording affected
women’s appraisals of the job, suggesting that the wording differ-
ences found at an institutional level in Studies 1 and 2 have the
ability to affect individuals in a way that maintains gender inequal-
ity in numerous fields. In Studies 3–5 jobs advertised with mas-
culine wording were seen as less gender diverse, and women found
those jobs less appealing, compared with jobs advertised with
feminine wording. This effect occurred, at least in part, because
women anticipated less belongingness in the positions advertised
with the masculinely worded descriptions (Studies 4 and 5). Ad-
ditionally, Study 5 demonstrated that gendered wording did not
affect people’s appraisals of their personal ability to carry out a
job. The effects of gendered wording, therefore, seem to operate
via changes in how much one anticipates belonging in a job and
less via an appraisal of whether one has the skill for that job.
Generally, gendered wording had the largest effect on women.
Men were only slightly more likely to find the masculinely worded
jobs more appealing than femininely worded jobs, and there was
no effect of gendered wording on men’s feelings of belongingness
within the occupation. The type of job participants read about (i.e.,
engineer, real estate broker, nurse) did not interact with the ma-
nipulations. Regardless of the type of job, participants, particularly
women, ranked jobs most highly when they included words that
matched their gender. Furthermore, not a single participant in
postexperimental debriefings suggested that his or her responses
were influenced by the wording of the advertisements or the extent
to which advertisements included words that conformed to gender
stereotypes.
Implications for Social Psychological Theories of
Inequality and Sexism
This research documents the existence of a structural mecha-
nism that reflects and reaffirms gender inequality, manifesting
subtly through the language employed in job advertisements, and
thus offers unique empirical support for previous theory, most
notably SDT (Sidanius & Pratto 1999). Of all social psychological
intergroup theories, SDT articulates most comprehensively how
group-based hierarchy—such as gender-based division of la-
6
To further rule out differences in perceived skill as an alternative
explanation, we let a separate sample of 73 participants (36 women, 37
men) read all six advertisements. Participants assessed the training and skill
required for each job, independent of their own ability to perform that job
(four items; e.g., “How much skill do you think this job requires?”; alphas
ranged from .70 to .94). There was no effect of wording on skill ratings (all
ps.22). Furthermore, after reading all six advertisements, participants
rated the appeal of each job using one item per job. The appeal effects
found in Studies 4 and 5 were replicated for female participants: We found
a significant Wording Gender interaction, F(1, 70) 5.17, p03,
p
2
.07. Women found femininely worded advertisements more appeal-
ing (M4.31, SE 0.19) than masculinely worded advertisements (M
3.89, SE 0.17), F(1, 35) 4.53, p.04. For men in this sample there
was no effect of wording, F(1, 35) 1.01, p.32.
Belongingness
61
**
-.19* .
61
Advertisement Wording Job Appeal
-.09
(-.21*)
Personal Skills
-.04 .15*
Personal
Skills
Figure 4. Standardized path estimates from Study 5 (all female partici-
pants). Advertisement wording coded as 1 masculine wording and 0
feminine wording.
p.05.
ⴱⴱ
p.001.
11
GENDERED WORDING AND INEQUALITY
bor—is maintained through institutional discrimination. Past em-
pirical support for the theory has focused on how institutional
discrimination manifests within specific contexts—such as the
criminal justice system (Sidanius, Liu, Shaw, & Pratto, 1994),
government policy (Pratto, Stallworth, & Conway-Lanz, 1998),
and corporations (Pratto & Espinoza, 2001)—and through
individual-mediated practices, such as when individuals, acting on
behalf of the state, allow their biased beliefs to influence decisions
or actions they take. The current analysis expands on this past
work by documenting an institutional-level barrier that does not
solely rely on individual-mediated practices but is ingrained di-
rectly into recruitment practices that perpetuate gender divisions.
The current work also demonstrates that SDT-type effects can
operate via mechanisms of belongingness.
Such institutional procedures (i.e., the way that job advertise-
ments are written) likely have a discriminatory effect even if there
is no discriminatory intent. To be clear, it is unlikely that gendered
wording is deliberately infused in advertisements. Unlike in the
1960s when employers advertised specifically for men, today
many companies have explicit directives, and genuine goals, to
gender diversify their workplaces. Thus, it is likely that most, if not
all, gendered wording emerges through motivational biases that
operate outside people’s awareness. Regardless of whether gen-
dered wording deliberately ends up in advertisements, the natural-
istic data suggest that it is common in male-dominated fields, and
contributes to the division of traditional gender roles by dissuading
women’s interest in jobs that are masculine worded.
Our experimental data also suggest that replacing the masculine
wording with parallel feminine wording would increase women’s
interest in those jobs. The consequences of adding feminine word-
ing to real job advertisements, however, may not be so straight-
forward and direct. There is some evidence that “feminizing” job
descriptions may implicitly introduce requirements for warmth or
communion and increase discrimination against agentic women,
who are seen as insufficiently “nice” (Rudman & Glick, 1999).
Adding feminine wording to a job advertisement may attract more
female applicants, but what if those applicants are now evaluated
differently by the people making hiring decisions? It would be
unfortunate if a femininely worded job advertisement generated
more job interest from women generally, but then agentic female
applicants were subsequently less likely to be hired. These con-
siderations make it especially important to systematically and
thoroughly investigate the role of gendered wording in the recruit-
ment and hiring process before making strong policy recommen-
dations.
Although Studies 1 and 2 suggest that social dominance pro-
cesses underlie the emergence of masculine wording in advertise-
ments, both SRT and SDT are useful for explaining gendered
divisions of work more generally. Masculine wording is a barrier
that qualified women face when exploring career options. How-
ever, consistent with SRT, widely held gender stereotypes likely
contribute to the fact that women, on average, are less likely than
men to undertake initial training in highly male-dominated areas.
Focusing on the institutional-level factors contributing to gendered
divisions of work highlights a point of intervention: If stereotypes
arise, at least in part, because of broad differences in role divisions,
as SRT proposes, then changing women’s position in the social
structure may change the stereotype associated with women (see
Diekman & Eagly, 2000). A stereotype of greater female agency
could increase women’s training in currently male-dominated do-
mains. Thus, eliminating the use of incidental masculine wording
in job advertisements may not only increase the numbers of
women in these occupations but change the female stereotype to
include more agentic traits, leading to greater numbers of women
seeking training in these occupations.
Also noteworthy is the subtlety with which structural contribu-
tors to inequality are woven into the system. The mere presence of
masculine words such as challenge,lead,boast, and active in
advertisements made jobs appear less gender diverse (Study 3) and
less appealing to women (Studies 4 and 5), even though these
words composed a fraction of the total words in the advertise-
ments. This subtlety of gendered wording makes it a particularly
pernicious and potent contributor to inequality. Whereas gendered
pronouns and other explicit references to the gender of a candidate
can be detected by readers, gendered wording, in the way that we
have studied it here, is masked. Not one participant picked up on
gendered language when asked at the end of the studies what
aspects of the advertisements affected his or her opinion about the
job. Instead, participants generally attributed their decisions to
broad attitudes toward the types of careers being advertised. Al-
though these attitudes certainly played a role in their responses,
given that participants were randomly assigned to experimental
condition, wording manipulations also mattered. Consistently find-
ing certain jobs slightly more unappealing over time, without being
aware of the external reasons why, may actually cause some
women to believe that they are uninterested, for intrinsic reasons,
in particular types of jobs. This may ultimately lead them away
from occupations that they may otherwise have found more intrin-
sically interesting.
There are clear implications of the present work for sex dis-
crimination literatures. To date, many of the well-documented
barriers to women’s inclusion in traditionally masculine occupa-
tions have been studied primarily in evaluation and performance
contexts—and after women have already been hired. Such better
documented barriers include (a) exclusion from informal network-
ing opportunities necessary for promotion (Ibarra, 1992), (b) un-
welcoming or patronizing work environments (Dardenne, Dumont,
& Bollier, 2007; Vescio, Gervais, Snyder, & Hoover, 2005), (c)
double standards for promotion and hiring (Biernat & Kobrynow-
icz, 1997; Cuddy et al., 2004; Rudman, 1998; Rudman & Glick,
1999), (d) attributional rationalizations (Heilman & Haynes,
2005), and (e) underperformance due to stereotype threat (Bell,
Spencer, Iserman, & Logel, 2003; Davies & Spencer, 2005). The
current work complements this past work by identifying a subtle
institutional-level barrier operating at the initial recruitment phase.
The current work also comments on self-selection hypotheses
and explanations for women’s career choices. What may appear, at
least at first glance, as gender differences in self-selection of
occupations is actually influenced, at least in part, by systematic
factors outside the self. It is not only that women are socialized to
like occupations that are thought to require highly feminine traits,
such as nursing, though this is certainly part of the puzzle of
women’s underrepresentation. Additionally, as the current work
reveals, subtle cues outside the self are affecting the appeal of
various jobs for women. Women and men, for example, may
equally like and desire an engineering job, but highly masculine
wording used in the advertisement reduces women’s appeal of the
job because it cues that women do not belong. Highly masculine
12 GAUCHER, FRIESEN, AND KAY
wording may cause women to turn away from a job that they may
have otherwise liked because they suspect that they will not
belong, not because they do not think that they have the skills
necessary for the job. In this way, qualified applicants are opting
out of jobs that they could perform well. We speculate that
women’s reaction to highly masculine wording is likely functional
and adaptive. Forced to overcome barriers to advancement and
inclusion within male-dominated occupations, women in these
fields have likely learned that certain environments are more
hostile toward women than others. Being vigilant for cues altering
them to the possibility of future negative experiences is important
for future occupational success.
Limitations and Future Directions
Although the laboratory is a good place to magnify novel
phenomena, laboratory experiments can raise concerns about ex-
ternal validity. Indeed, the extent to which the stimulus materials
used in Studies 3–5 map onto the real-world findings of Studies 1
and 2 is an issue that is important for evaluating the magnitude of
the current work. Certainly the average presence of masculine and
feminine words in the controlled experimental materials was
higher than the average emerging in the real-world advertisements,
yet the written experimental advertisements used in Studies 3–5
were not completely out of bounds according to the real-world
advertisements. The created advertisements contained an average
of approximately 7%– 8% gendered language, whereas the real-
world advertisements contained an average of approximately 1%
gendered language. Examining the range of the real-world sample,
however, demonstrates that real-word advertisements can contain
up to 8% gendered language. Even though such advertisements
may be outliers in these samples, they do exist.
Also, strong manipulations containing a high presence of gen-
dered wording were necessary because experimental participants
had read only one or six advertisements. Actual job seekers typi-
cally read dozens, maybe hundreds, of advertisements. Thus the
“dose” of gendered language is comparable, whether it is received
as a strong bias across six advertisements or a less strong bias
across 50 advertisements. That said, future work should test the
boundary conditions of gendered wording effects and determine
the smallest amount of masculine wording necessary to affect
individual women. Furthermore, although the type of wording
found in male-dominated areas (many masculine words, few fem-
inine words) was replicated in Studies 3–5, the type of wording
found in female-dominated areas (moderate levels of both mascu-
line and feminine words) was not exactly represented in our
laboratory studies. In the current work we chose to focus on how
women react to the masculine wording typically found in male-
dominated fields, and whether they might find advertisements
more appealing if they were femininely worded, but future work
should investigate reactions to the more balanced wording typi-
cally found in female-dominated areas. Future research should also
identify the important features of masculine wording in determin-
ing job appeal (or other effects). For example, the location of
gendered wording within an advertisement may moderate its ef-
fect. Gendered wording within the company description (e.g., “We
are a dominant firm”) might affect global perceptions of the
company as a whole, whereas gendered wording within the de-
scription of job duties (e.g., “You will be a dominant sales force”)
might affect only perceptions of the job, not perceptions of the
company more globally. A limitation of the current studies is that
gendered wording was diffused throughout each advertisement and
not designed to test whether there might be different effects in
different parts of the job advertisement. Also, the current studies
did not examine precisely which words were most powerful in
driving gendered wording effects. It is likely that words most
strongly associated with the male stereotype will have the largest
effect than words more weakly associated with the male stereo-
type.
To understand why women continue to be underrepresented in
traditionally male-dominated fields, it is essential to understand
precisely how institutional-level factors, such as gendered word-
ing, affect the appeal of certain jobs. Understanding institutional-
level factors is also important if one wants to design initiatives or
interventions that will increase gender diversity, or diversity of all
types. There are certainly groups within other domains (such as
universities) that would like to diversify their membership. For
example, the wording of some recruitment materials, such as
university mission statements, may tacitly appeal to people of
different socioeconomic status.
Just as gendered wording is salient to men and women, wording
associated with individualism– collectivism may be salient to peo-
ple of high and low socioeconomic status. Individualism—the
degree to which an autonomous and unique self is valued over the
group—is most strongly a characteristic of the White, urban, upper
middle class (Oyserman & Marcus, 1998). As such, words asso-
ciated with individualism (such as individual and uniqueness) may
be more likely to emerge within the mission statements of high-
status or private universities, compared with the mission state-
ments of lower status or public universities. Just as highly mascu-
line wording makes women feel like they do not belong, highly
individualistic wording within the mission statements of private
universities may subtly cue that members of lower socioeconomic
status groups do not belong and discourage them from applying,
apart from other tangible reasons (e.g., tuition costs). Thus, our
focus on how language perpetuates existing group dominance has
theoretical significance for a host of other domains, not only for
gender and occupations. Future work would do well to explore
how language manifesting at the institutional-level inadvertently
serves to maintain existing social arrangements across a number of
domains.
Finally, there are a few limitations of the current work that are
worth noting. Students from a major Canadian university, although
from a broad range of academic majors, were participants in
Studies 3–5. Although in one sense university students are relevant
participants insofar as they do not currently have a job and are all
anticipating applying for one in the near future, one may wonder
whether the current findings would generalize to other demo-
graphic groups or to active job applicants. Active job applicants,
for instance, may be more highly committed, or motivated, to get
a job in a particular occupation. On the one hand, a high level of
motivation and commitment could make people more attentive to
wording cues, in an effort to find the perfect job. On the other
hand, strong motivation and commitment to a specific occupation
could cause people to apply indiscriminately to jobs in an attempt
to ensure that they get at least some job in the field, thus reducing
the strength of gendered wording on people’s appeal judgments.
13
GENDERED WORDING AND INEQUALITY
Elucidating such moderators of gendered wording effects is a
fruitful future direction.
It is also important to note the nature of our dependent variable.
Job appeal was assessed with a six-item measure, showing good
reliability across studies. This measure was reliable for both men
and women, and can be taken as face valid index of the extent to
which a job is appealing. Nonetheless, future work would do well
to include a true behavioral measure.
Conclusion
Social inequality and existing divisions of labor can be main-
tained and perpetuated in many ways. Individuals and groups hold
a range of ideologies, belief systems, and stereotypes that justify
the status quo (Jost & Hunyady, 2005; Napier, Mandisodza, An-
dersen, & Jost, 2006), but individuals need not always be the
driving force. Inequalities can also be reinforced via institutional-
level factors that influence individual judgments and preferences
in such ways that serve to preserve group inequality and the
prevailing status quo. Although these institutional-level features
may be less salient or seemingly direct than particular justifying or
hierarchy-enhancing ideologies, and are much less researched,
they are likely just as important in stagnating social change (Haley
& Sidanius, 2005). The current research highlights one such
institutional-level feature and demonstrates its potential impact on
judgments relevant to the maintenance of inequality. In doing so,
it provides useful advances for our understanding of gender in-
equality in the workforce. But beyond that, we hope it highlights
the power of looking to features of the social structure— especially
those that may easily go unnoticed—in helping social psycholo-
gists uncover the ways in which social inequality is created,
reinforced, and ultimately maintained.
References
Adams, G., Biernat, M., Branscombe, N. R., Crandall, C. S., & Wrights-
man, L. S. (2008). Beyond prejudice: Toward a sociocultural psychology
of racism and oppression. In Commemorating Brown: The social psy-
chology of racism and discrimination (pp. 215–246). Washington, DC:
American Psychological Association. doi:10.1037/11681-012
Bartz, J. A., & Lydon, J. E. (2004). Close relationships and the working
self-concept: Implicit and explicit effects of priming attachment on
agency and communion. Personality and Social Psychology Bulletin, 30,
1389 –1401. doi:10.1177/0146167204264245
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for
interpersonal attachments as a fundamental human motivation. Psycho-
logical Bulletin, 117, 497–529. doi:10.1037/0033-2909.117.3.497
Bell, A. E., Spencer, S. J., Iserman, E., & Logel, C. E. R. (2003). The effect
of stereotype threat on women’s engineering performance. Journal of
Engineering Education, 92, 307–312.
Bem, S. L. (1974). The measurement of psychological androgyny. Journal
of Consulting and Clinical Psychology, 42, 155–162. doi:10.1037/
h0036215
Bem, S. J., & Bem, D. J. (1973). Does sex-biased job advertising “aid and
abet” sex discrimination? Journal of Applied Social Psychology, 3,
6 –18. doi:10.1111/j.1559-1816.1973.tb01290.x
Biernat, M., & Kobrynowicz, D. (1997). Gender- and race-based standards
of competence: Lower minimum standards but higher ability standards
for devalued groups. Journal of Personality and Social Psychology, 72,
544 –557. doi:10.1037/0022-3514.72.3.544
Boroditsky, L. (2001). Does language shape thought? Mandarin and Eng-
lish speakers’ conceptions of time. Cognitive Psychology, 43, 1–22.
doi:10.1006/cogp.2001.0748
Brownlow, S., Rosamond, J. A., & Parker, J. A. (2003). Gender-linked
linguistic behavior in television interviews. Sex Roles, 49, 121–132.
doi:10.1023/A:1024404812972
Carli, L. L. (1990). Gender, language, and influence. Journal of Person-
ality and Social Psychology, 59, 941–951. doi:10.1037/0022-
3514.59.5.941
Catalyst. (2008a). U.S. women in business. Retrieved from http://
www.catalyst.org/publication/132/us-women-in-business
Catalyst. (2008b). Women in the sciences. Retrieved from http://
www.catalyst.org/file/205/qt_women_in_sciences.pdf
Cheryan, S., Plaut, V. C., Davis, P. G., & Steele, C. M. (2009). Ambient
belonging: How stereotypical cues impact gender participation in com-
puter science. Journal of Personality and Social Psychology, 97, 1045–
1060. doi:10.1037/a0016239
Cuddy, A. J. C., Fiske, S. T., & Glick, P. (2004). When professionals
become mothers, warmth doesn’t cut the ice. Journal of Social Issues,
60, 701–718. doi:10.1111/j.0022-4537.2004.00381.x
Dardenne, B., Dumont, M., & Bollier, T. (2007). Insidious dangers of
benevolent sexism: Consequences for women’s performance. Journal of
Personality and Social Psychology, 93, 764 –779. doi:10.1037/0022-
3514.93.5.764
Davies, P. G., & Spencer, S. J. (2005). The gender-gap artifact: Women’s
underperformance in quantitative domains through the lens of stereotype
threat. In A. M. Gallagher & J. C. Kaufman (Eds.), Gender differences
in mathematics: An integrative psychological approach (pp. 172–188).
New York, NY: Cambridge University Press.
Davies, P. G., Spencer, S. J., Quinn, D. M., & Gerhardstein, R. (2002).
Consuming images: How television commercials that elicit stereotype
threat can restrain women academically and professionally. Personality
and Social Psychology Bulletin, 28, 1615–1628. doi:10.1177/
014616702237644
Deutsch, M. (2006). A framework for thinking about oppression and its
change. Social Justice Research, 19, 7– 41. doi:10.1007/s11211-006-
9998-3
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. doi:10.1177/0146167200262001
Eagly, A. H. (1987). Sex differences in social behavior: A social-role
interpretation. Hillsdale, NJ: Erlbaum.
Eagly, A. H., & Karau, S. J. (1991). Gender and the emergence of leaders:
A meta-analysis. Journal of Personality and Social Psychology, 60,
685–710. doi:10.1037/0022-3514.60.5.685
Engineers Canada. (2010). Women in engineering. Retrieved from http://
www.engineerscanada.ca/e/pr_women.cfm
Fiske, S. T., Bersoff, D. N., Borgida, E., Deaux, K., & Heilman, M. E.
(1991). Social science research on trial: Use of sex stereotyping research
in Price Waterhouse v. Hopkins.American Psychologist, 46, 1049 –
1060. doi:10.1037/0003-066X.46.10.1049
Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). A model of (often
mixed) stereotype content: Competence and warmth respectively follow
from perceived status and competition. Journal of Personality and
Social Psychology, 82, 878 –902. doi:10.1037/0022-3514.82.6.878
Fitzsimons, G. M., & Kay, A. C. (2004). Language and interpersonal
cognition: Causal effects of variations in pronoun usage on perceptions
of closeness. Personality and Social Psychology Bulletin, 30, 547–557.
doi:10.1177/0146167203262852
Glick, P., & Fiske, S. T. (1996). The Ambivalent Sexism Inventory:
Differentiating hostile and benevolent sexism. Journal of Personality
and Social Psychology, 70, 491–512. doi:10.1037/0022-3514.70.3.491
Glick, P., & Fiske, S. T. (2001a). An ambivalent alliance: Hostile and
benevolent sexism as complementary justifications for gender inequal-
14 GAUCHER, FRIESEN, AND KAY
ity. American Psychologist, 56, 109 –118. doi:10.1037/0003-
066X.56.2.109
Glick, P., & Fiske, S. T. (2001b). Ambivalent stereotypes as legitimizing
ideologies: Differentiating paternalistic and envious prejudice. In J. T.
Jost & B. Major (Eds.), The psychology of legitimacy: Emerging per-
spectives on ideology, justice, and intergroup relations (pp. 278 –306).
New York, NY: Cambridge University Press.
Haas, A. (1979). Male and female spoken language differences: Stereo-
types and evidence. Psychological Bulletin, 86, 616 – 626. doi:10.1037/
0033-2909.86.3.616
Haley, H., & Sidanius, J. (2005). Person– organization congruence and the
maintenance of group-based social hierarchy: A social dominance per-
spective. Group Processes & Intergroup Relations, 8, 187–203. doi:
10.1177/1368430205051067
Heilman, M. E. (1983). Sex bias in work settings: The lack of fit model.
Research in Organizational Behavior, 5, 269 –298.
Heilman, M. E., & Haynes, M. C. (2005). No credit where credit is due:
Attributional rationalization of women’s success in male–female teams.
Journal of Applied Psychology, 90, 905–916. doi:10.1037/0021-
9010.90.5.905
Hoffman, C., & Hurst, N. (1990). Gender stereotypes: Perception or
rationalization? Journal of Personality and Social Psychology, 58, 197–
208. doi:10.1037/0022-3514.58.2.197
Hoffman, C., & Tchir, M. A. (1990). Interpersonal verbs and dispositional
adjectives: The psychology of causality embodied in language. Journal
of Personality and Social Psychology, 58, 765–778. doi:10.1037/0022-
3514.58.5.765
Holmes, M. A., & O’Connell, S. (2007). Leaks in the pipeline. Nature, 446,
346. doi:10.1038/nj7133–346a
Ibarra, H. (1992). Homophily and differential returns: Sex differences in
network structure and access in an advertising firm. Administrative
Science Quarterly, 37, 422– 447. doi:10.2307/2393451
Jost, J. T., & Banaji, M. R. (1994). The role of stereotyping in system-
justification and the production of false consciousness. British Journal of
Social Psychology, 33, 1–27.
Jost, J. T., Banaji, M. R., & Nosek, B. A. (2004). A decade of system
justification theory: Accumulated evidence of conscious and uncon-
scious bolstering of the status quo. Political Psychology, 25, 881–919.
doi:10.1111/j.1467-9221.2004.00402.x
Jost, J. T., & Hunyady, O. (2005). Antecedents and consequences of
system-justifying ideologies. Current Directions in Psychological Sci-
ence, 14, 260 –265. doi:10.1111/j.0963-7214.2005.00377.x
Jost, J. T., & Kay, A. C. (2005). Exposure to benevolent sexism and
complementary gender stereotypes: Consequences for specific and dif-
fuse forms of system justification. Journal of Personality and Social
Psychology, 88, 498 –509. doi:10.1037/0022-3514.88.3.498
Judd, C. M., Kenny, D. A., & McClelland, G. H. (2001). Estimating and
testing mediation and moderation in within-subject designs. Psycholog-
ical Methods, 6, 115–134. doi:10.1037/1082-989X.6.2.115
Kay, A. C., Czaplin´ ski, S., & Jost, J. T. (2009). Left–right ideological
differences in system justification following exposure to complementary
versus noncomplementary stereotype exemplars. European Journal of
Social Psychology, 39, 290 –298. doi:10.1002/ejsp.500
Kay, A. C., Gaucher, D., Peach, J. M., Laurin, K., Friesen, J., Zanna, M. P.,
& Spencer, S. J. (2009). Inequality, discrimination, and the power of the
status quo: Direct evidence for a motivation to see the way things are as
the way they should be. Journal of Personality and Social Psychology,
97, 421– 434. doi:10.1037/a0015997
Kay, A. C., Jost, J. T., Mandisodza, A. N., Sherman, S. J., Petrocelli, J. V.,
& Johnson, A. L. (2007). Panglossian ideology in the service of system
justification: How complementary stereotypes help us to rationalize
inequality. In M. P. Zanna (Ed.), Advances in experimental social
psychology (Vol. 39, pp. 305–358). San Diego, CA: Elsevier. doi:
10.1016/S0065-2601(06)39006-5
Kay, A. C., & Zanna, M. P. (2009). A contextual analysis of the system
justification motive and its societal consequences. In J. T. Jost, A. C.
Kay, & H. Thorisdottir (Eds.), Social and psychological bases of ideol-
ogy and system justification (pp. 158 –181). New York, NY: Oxford
University Press. doi:10.1093/acprof:oso/9780195320916.003.007
Kervyn, N., Yzerbyt, V. Y., Judd, C. M., & Nunes, A. (2009). A question
of compensation: The social life of the fundamental dimensions of social
perception. Journal of Personality and Social Psychology, 96, 828 – 842.
doi:10.1037/a0013320
Lakoff, R. T. (1975). Language and woman’s place. New York, NY:
Harper & Row.
Leaper, C., & Ayres, M. M. (2007). A meta-analytic review of gender
variations in adults’ language use: Talkativeness, affiliative speech, and
assertive speech. Personality and Social Psychology Review, 11, 328 –
363. doi:10.1177/1088868307302221
Maass, A. (1999). Linguistic intergroup bias: Stereotype perpetuation
through language. In M. P. Zanna (Ed.), Advances in experimental social
psychology (Vol. 31, pp. 79 –121). San Diego, CA: Academic Press.
doi:10.1016/S0065-2601(08)60272-5
Maass, A., Salvi, D., Arcuri, L., & Semin, G. (1989). Language use in
intergroup contexts: The linguistic intergroup bias. Journal of Person-
ality and Social Psychology, 57, 981–993. doi:10.1037/0022-
3514.57.6.981
MacKinnon, D. P., Lockwood, C. M., & Williams, J. (2004). Confidence
limits for the indirect effect: Distribution of the product and resampling
methods. Multivariate Behavioral Research, 39, 99 –128. doi:10.1207/
s15327906mbr3901_4
Madera, J. M., Hebl, M. R., & Martin, R. C. (2009). Gender and letters of
recommendation for academia: Agentic and communal differences.
Journal of Applied Psychology, 94, 1591–1599. doi:10.1037/a0016539
Murphy, M. C., Steele, C. M., & Gross, J. J. (2007). Signaling threat: How
situational cues affect women in math, science, and engineering. Psy-
chological Science, 18, 879 – 885. doi:10.1111/j.1467-9280.2007.
01995.x
Napier, J. L., Mandisodza, A. N., Andersen, S. M., & Jost, J. T. (2006).
System justification in responding to the poor and displaced in the
aftermath of Hurricane Katrina. Analyses of Social Issues and Public
Policy, 6, 57–73. doi:10.1111/j.1530-2415.2006.00102.x
Napier, J. L., Thorisdottir, H., & Jost, J. T. (2010). The joy of sexism? A
multinational investigation of hostile and benevolent justifications for
gender inequality and their relations to subjective well-being. Sex Roles,
62, 405– 419. doi:10.1007/s11199-009-9712-7
Newcombe, N., & Arnkoff, D. B. (1979). Effects of speech style and sex
of speaker on person perception. Journal of Personality and Social
Psychology, 37, 1293–1303. doi:10.1037/0022-3514.37.8.1293
Newman, M. L., Groom, C. J., Handelman, L. D., & Pennebaker, J. W. (2008).
Gender differences in language use: An analysis of 14,000 text samples.
Discourse Processes, 45, 211–236. doi:10.1080/01638530802073712
Oyserman, D., & Markus, H. R. (1998). Self as social representation. In U.
Flick (Ed.), The psychology of the social (pp. 107–125). New York, NY:
Cambridge University Press.
Pedriana, N., & Abraham, A. (2006). Now you see them, now you don’t:
The legal field and newspaper desegregation of sex-segregated help
wanted ads 1965–75. Law & Social Inquiry, 31, 905–938. doi:10.1111/
j.1747-4469.2006.00039.x
Pennebaker, J. W., Booth, R. J., & Francis, M. E. (2007). Linguistic
Inquiry and Word Count: LIWC 2007 [Computer software]. Austin, TX:
LIWC.
Pennebaker, J. W., Mehl, M. R., & Niederhoffer, K. G. (2003). Psycho-
logical aspects of natural language use: Our words, our selves. Annual
Review of Psychology, 54, 547–577. doi:10.1146/annurev.
psych.54.101601.145041
Phelan, J. E., Moss-Racusin, C. A., & Rudman, L. A. (2008). Competent
yet out in the cold: Shifting criteria for hiring reflect backlash toward
15
GENDERED WORDING AND INEQUALITY
agentic women. Psychology of Women Quarterly, 32, 406 – 413. doi:
10.1111/j.1471-6402.2008.00454.x
Pratto, F., & Espinoza, P. (2001). Gender, ethnicity, and power. Journal of
Social Issues, 57, 763–780. doi:10.1111/0022-4537.00240
Pratto, F., Sidanius, J., & Levin, S. (2006). Social dominance theory and
the dynamics of intergroup relations: Taking stock and looking forward.
European Review of Social Psychology, 17, 271–320. doi:10.1080/
10463280601055772
Pratto, F., Stallworth, L. M., & Conway-Lanz, S. (1998). Social dominance
orientation and the ideological legitimization of social policy. Journal of
Applied Social Psychology, 28, 1853–1875. doi:10.1111/j.1559-
1816.1998.tb01349.x
Purdie-Vaughns, V., Steele, C. M., Davies, P. G., Ditlmann, R., & Crosby,
J. R. (2008). Social identity contingencies: How diversity cues signal
threat or safety for African Americans in mainstream institutions. Jour-
nal of Personality and Social Psychology, 94, 615– 630. doi:10.1037/
0022-3514.94.4.615
Reitsma-van Rooijen, M., Semin, G. R., & van Leeuwen, E. (2007). The
effect of linguistic abstraction on interpersonal distance. European Jour-
nal of Social Psychology, 37, 817– 823. doi:10.1002/ejsp.410
Rudman, L. A. (1998). Self-promotion as a risk factor for women: The
costs and benefits of counterstereotypical impression management. Jour-
nal of Personality and Social Psychology, 74, 629 – 645. doi:10.1037/
0022-3514.74.3.629
Rudman, L. A., & Glick, P. (1999). Feminized management and backlash
toward agentic women: The hidden costs to women of a kinder, gentler
image of middle managers. Journal of Personality and Social Psychol-
ogy, 77, 1004 –1010. doi:10.1037/0022-3514.77.5.1004
Rudman, L. A., & Glick, P. (2001). Prescriptive gender stereotypes and
backlash toward agentic women. Journal of Social Issues, 57, 743–762.
doi:10.1111/0022-4537.00239
Rudman, L. A., & Kilianski, S. T. (2000). Implicit and explicit attitudes
toward female authority. Personality and Social Psychology Bulletin, 26,
1315–1328. doi:10.1177/0146167200263001
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the
facilitation of intrinsic motivation, social development, and well-being.
American Psychologist, 55, 68 –78. doi:10.1037/0003-066X.55.1.68
Sadler, P., Ethier, N., & Woody, E. (in press). Tracing the interpersonal
web of psychopathology: Dyadic data analysis methods for clinical
researchers. Journal of Experimental Psychopathology.
Schmader, T., Whitehead, J., & Wysocki, V. H. (2007). A linguistic
comparison of letters of recommendation for male and female chemistry
and biochemistry job applicants. Sex Roles, 57, 509 –514. doi:10.1007/
s11199-007-9291-4
Schullo, S. A., & Alperson, B. L. (1984). Interpersonal phenomenology as
a function of sexual orientation, sex, sentiment, and trait categories in
long-term dyadic relationships. Journal of Personality and Social Psy-
chology, 47, 983–1002. doi:10.1037/0022-3514.47.5.983
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and non-
experimental studies: New procedures and recommendations. Psycho-
logical Methods, 7, 422– 445. doi:10.1037/1082-989X.7.4.422
Sidanius, J., Liu, J. H., Shaw, J. S., & Pratto, F. (1994). Social dominance
orientation, hierarchy attenuators and hierarchy enhancers: Social dom-
inance theory and the criminal justice system. Journal of Applied Social
Psychology, 24, 338 –366. doi:10.1111/j.1559-1816.1994.tb00586.x
Sidanius, J., & Pratto, F. (1999). Social dominance: An intergroup theory
of social hierarchy and oppression. New York, NY: Cambridge Uni-
versity Press.
Tesch, B. J., Wood, H. M., Helwig, A. L., & Nattinger, A. B. (1995).
Promotion of women physicians in academic medicine: Glass ceiling or
sticky floor? Journal of the American Medical Association, 273, 1022–
1025. doi:10.1001/jama.273.13.1022
U.S. Department of Labor. (2007). Median weekly earnings of full-time
wage and salary workers by detailed occupation and sex. Retrieved from
http://www.bls.gov/cps/cpsaat39.pdf
Vescio, T. K., Gervais, S. J., Snyder, M., & Hoover, A. (2005). Power and
the creation of patronizing environments: The stereotype-based behav-
iors of the powerful and their effects on female performance in mascu-
line domains. Journal of Personality and Social Psychology, 88, 658 –
672. doi:10.1037/0022-3514.88.4.658
Walton, G. M., & Cohen, G. L. (2007). A question of belonging: Race,
social fit, and achievement. Journal of Personality and Social Psychol-
ogy, 92, 82–96. doi:10.1037/0022-3514.92.1.82
(Appendices follow)
16 GAUCHER, FRIESEN, AND KAY
Appendix A
List of Masculine and Feminine Words Coded in Studies 1 and 2
Masculine words Feminine words
Active Affectionate
Adventurous Child
AggressCheer
AmbitioCommit
AnalyCommunal
AssertCompassion
AthletConnect
AutonomConsiderate
BoastCooperat
ChallengDepend
CompetEmotiona
Confident Empath
CouragFeminine
Decide Flatterable
Decisive Gentle
DecisionHonest
DeterminInterpersonal
Dominant Interdependen
DominaInterpersona
ForceKind
Greedy Kinship
Headstrong Loyal
HierarchModesty
HostilNag
Implusive Nurtur
IndependenPleasant
IndividualPolite
IntellectQuiet
LeadRespon
Logic Sensitiv
Masculine Submissive
Objective Support
Opinion Sympath
Outspoken Tender
Persist Together
PrincipleTrust
Reckless Understand
Stubborn Warm
Superior Whin
Self-confidenYield
Self-sufficien
Self-relian
Note. The asterisk denotes the acceptance of all letters, hyphens, or numbers following its appearance.
(Appendices continue)
17
GENDERED WORDING AND INEQUALITY
Appendix B
Job Advertisements Used in Studies 3–5
Feminine Masculine
Engineer
Company description Company description
We are a community of engineers who have effective
relationships with many satisfied clients.
We are a dominant engineering firm that boasts many
leading clients.
We are committed to understanding the engineering
sector intimately.
We are determined to stand apart from the
competition.
Qualifications Qualifications
Proficient oral and written communication skills. Strong communication and influencing skills.
Collaborates well, in a team environment.
Sensitive to clients’ needs, can develop warm client
relationships.
Bachelor of Engineering degree or higher from
recognized university.
Registered as a Professional Engineer.
Ability to perform individually in a competitive
environment.
Superior ability to satisfy customers and manage
company’s association with them.
Bachelor of Engineering degree or higher from
recognized university.
Registered as a Professional Engineer.
Responsibilities Responsibilities
Provide general support to project teams in a manner
complimentary to the company.
Direct project groups to manage project progress and
ensure accurate task control.
Help clients with construction activities. Determine compliance with client’s objectives.
Create quality engineering designs. Create quality engineering designs.
Plumber
Company description Company description
We are a committed provider of dependable plumbing
solutions.
We are determined company that delivers superior
plumbing.
We have many loyal clients, and deliver honest,
personal service.
We are proud of our success, and boast an impressive
record.
Qualifications Qualifications
Dependable, with demonstrated commitment to client
service.
Self-reliant, with demonstrated ability to perform
tasks independently.
Can interpret blueprints and schematics. Ability to analyze blueprints and schematics.
Licensed/certified plumber from recognized
community college or related program.
Licensed/certified plumber from recognized
community college or related program.
Previous experience an asset. Previous experience an asset.
Responsibilities Responsibilities
Service our clients’ plumbing systems. Maintain customers’ plumbing systems.
Respond to plumbing problems and find innovative
repair solutions.
Analyze problems logically and troubleshoot to
determine needed repairs.
Collaborate on new building projects, providing
plumbing advice.
Deliver plumbing expertise on new building projects.
Retail sales manager
Company description Company description
Our hope is to be the best employer in clothing retail
by providing a pleasant and rewarding employment
experience.
Our ambition is to be the best employer in clothing
retail by delivering a rewarding employment
experience.
We nurture and support our employees, expecting that
they will become committed to their chosen career.
We will challenge our employees to be proud of their
chosen career.
Qualifications Qualifications
Full-time, flexible availability. Full-time, variable availability.
Cheerful, with excellent communication skills. Strong communication skills.
Capable of working with minimal supervision. Ability to work independently.
(Appendices continue)
18 GAUCHER, FRIESEN, AND KAY
Appendix B (continued)
Feminine Masculine
Responsibilities Responsibilities
Can motivate others to reach their potential as
employees.
Can challenge others to reach their potential as
employees.
You will be the head of our fast-paced store, with
further opportunities for career development.
You will be the boss of our fast-paced store, with
further opportunities career advancement.
You’ll develop interpersonal skills and understanding
of business.
You’ll develop leadership skills and learn business
principles.
Be a role model for your store, representing our
exclusive brand.
Be a leader in your store, representing our exclusive
brand.
Real estate agent
Company description Company description
Join our sales community! Even if you have no
previous experience, we will help nurture and develop
your sales talents.
Take our sales challenge! Even if you have no
previous experience, we will facilitate the acquisition
of your sales abilities.
We support our employees with an excellent
compensation package.
We boast a competitive compensation package.
Qualifications Qualifications
As the ideal candidate, you will have a pleasant
attitude, dependable judgment, and be attentive to
details.
The superior candidate will have a self-confident
attitude, decisive judgment, and be detail-oriented.
Excellent communicator. Strong communicator.
Bilingualism is an asset. Bilingualism is an asset.
Previous background in real estate an asset, but not
required.
Previous background in real estate an asset, but not
required.
Responsibilities Responsibilities
Connect with potential clients; being sensitive to their
needs, introduce them to properties.
Recruit potential buyers; determine their interests and
lead them to properties.
Serve as the connection between your client and each
property’s seller.
Negotiate for your buyer with each property’s seller.
Understand real estate markets to establish appropriate
selling prices for properties.
Analyze real estate markets to determine appropriate
selling prices for properties.
Nurse
Company description Company description
Our hospital serves a community of approximately
100,000 people.
The hospital is located within a city of approximately
100,000 people.
We are committed to provide top quality health care
that considers the best needs of our patients.
We are determined to deliver superior medical
treatment tailored to each individual patient.
Qualifications Qualifications
Demonstrated ability to respond appropriately
emergencies.
Demonstrated ability to act decisively in emergencies.
Principled and self-reliant professional with
dedication to the profession.
Able to work shift work.
Bachelor of Nursing from recognized university.
Current license with College of Nurses in Ontario as
a Registered Nurse.
Caring and compassionate professional with a passion
for the profession.
Dependable and willing to work shift work.
Bachelor of nursing from recognized university.
Current license with College of Nurses in Ontario as a
Registered Nurse.
Responsibilities Responsibilities
As part of a team, provide health care services to the
wide variety of community members who come for
care.
In keeping with the hospital’s hierarchy, administer
medical interventions to the wide variety of
individuals who require treatment.
Consider patient symptoms in order to select
appropriate treatment and support.
Analyze patient symptoms to determine appropriate
interventions.
Collaborate with other care providers, but also able to
work independently.
Be self-reliant and able work independently, but also
able to work on a team.
(Appendices continue)
19
GENDERED WORDING AND INEQUALITY
Appendix B (continued)
Feminine Masculine
Administrative assistant
Company description Company description
We are a corporate team dedicated to supporting our
financial clients.
Our organization works to offer every possible
advantage to our banking customers.
Our company is devoted to providing a great work/life
balance and compensation package.
The company boasts impressive salaries, allowing our
employees with financial independence.
Qualifications Qualifications
Experience in providing administrative support
services to a team of senior management and
understand office organization.
Ability to deal with multiple senior staff in a
demanding environment and navigate office
hierarchy.
Polite; sensitive to needs of other employees and
clients.
Civil personality; aware of other workers’ and
customers’ requirements.
Dependable and responsible. Independent and self-reliant.
Capable computer skills. Strong computer skills.
Responsibilities Responsibilities
Support office team and assist with departmental
procedures so that work progresses more efficiently.
Organize and monitor office tasks and processes so
that work progresses more efficiently.
Connect and develop relationships with a variety of
clients (e.g., other businesses).
Able to interface with external parties (e.g., other
businesses).
Coordinate incoming and outgoing shipments. Control incoming and outgoing shipments.
Received March 31, 2010
Revision received October 13, 2010
Accepted October 15, 2010
20 GAUCHER, FRIESEN, AND KAY
... One of the most common reasons for exclusion risks in job ads is on behalf of gender [5][6][7]. Research has demonstrated the presence of gender stereotypes in job postings [8,9], confirming a congruence of linguistic manifestations of gender stereotypes with gender dominance of the respective positions. Gender stereotypes are generalized assumptions or preconceptions that affect behaviors and judgments which can influence others' evaluation of work-related skills [10] or career choices [11] but also one's own self-perception [12][13][14]. ...
... Gender stereotypes are generalized assumptions or preconceptions that affect behaviors and judgments which can influence others' evaluation of work-related skills [10] or career choices [11] but also one's own self-perception [12][13][14]. In the context of job applications, gender-stereotypical linguistic constructions can influence how respondents interpret and react to job ads [8,[15][16][17]. In particular, research indicates that women (in comparison to men) are more willing to apply on behalf of job profiles containing prototypically feminine traits [8,[16][17][18]. ...
... In the context of job applications, gender-stereotypical linguistic constructions can influence how respondents interpret and react to job ads [8,[15][16][17]. In particular, research indicates that women (in comparison to men) are more willing to apply on behalf of job profiles containing prototypically feminine traits [8,[16][17][18]. This is consistent with the 'lack of fit' model by Heilman [19,20] by which women, when comparing their own characteristics and competencies with job requirements, might regard occupations described with stereotypically male characteristics less attractive due to the incongruity with self-perception [18,21,22]. ...
Article
Full-text available
Job advertisements are often worded in ways that might pose discrimination risks leading to the exclusion of certain groups of applicants, particularly in relation to their gender. Especially in male-dominated professions or leadership roles, the specific linguistic formulation of job postings acquires relevance if more women are to be attracted to apply. Various technologies have emerged that offer automated text screening, some of them even suggesting alternative formulations to increase gender inclusivity. In this study we analyze four software providers on the German market using a corpus of ∼160, 000 job ads from three different platforms. We identify the relevant social psychological research on gender and language that is at the scientific core of these technologies. We show that, despite sharing a common foundation, the four tools assess the potential for exclusion in job postings in a considerably divergent way on multiple levels of comparison. We discuss the levers in the software pipeline of all four technologies, as well as the potential effect of certain implementation decisions, such as string-based vs. semantic approaches to computational processing of natural language. We argue that the ‘technological translation’ of research is extremely involved and further studies of its use in practice are needed to assess the potential for more gender equality.
... Job appeal is assessed as an indicator of desire and interest for a particular job (Gaucher et al., 2011), and is associated with perceptions of fit (Tresh et al., 2019) and intentions to apply for a job (Horvath & Sczesny, 2014). Furthermore, to confirm that drivers would be willing and open to the alternate occupations we also measured drivers' explicit interest in these jobs as alternatives to truck driving. ...
... Job Appeal. The appeal of each occupation was assessed with 4 items adapted from a 6item scale (Gaucher et al., 2011). Two items that referred to one's organization were not included. ...
Article
Automation continues to be a disruptive force in the workforce. In particular, new automated technologies are projected to replace many mid-skill jobs, potentially displacing millions of workers. Career planning agencies and other organizations can help support workers if they are able to effectively identify optimal transition occupations for displaced workers. We drew upon the 24.2 Occupational Information Network (O*NET) Database to conduct two related studies that identify alternate occupations for truck drivers, who are at risk of job loss due to the adoption of autonomous vehicles. In Study 1, we statistically compared the jobs that we identified based on different search methods using O*NET classifications based on their similarity to the knowledge, skills, values, and interests held by truck drivers. In Study 2, we conducted a survey of truck drivers to evaluate their perceptions of the occupations identified as objectively similar to their occupation. Results indicate that optimal transition occupations may be identified by searching for occupations that share skills as well as the same work activities/industry as a given occupation. These findings hold further implications for career planning organizations and policymakers to ease workforce disruption due to automation.
...  Lack of skilled workforce (e.g.: in the IT sector).  Lack of equal opportunities in the workplace (McLaughlin et al., 2004, Lane et al., 2015, Gaucher et al., 2011, Quillian et al., 2017.  Different attitude to fulfilment of tasks (Brands-Kehre, 2014). ...
Article
Full-text available
This article discusses the potential of the application of interactive micro-learning content that aims to develop intercultural competences in the workplace, focusing on the pedagogically-informed design of playful interactive learning resources for engaging learners with topics related to cultural risks. Eight topics were identified through our engagement with 154 personnel from multicultural organisations across five European countries, including cross-cultural awareness, stereotypes, and communication. A competence map was then developed, articulating the specific aspects of these topics, which informed online content development. Findings from the evaluation of the online educational materials across the five countries (n=223 participants) suggest the significance and relevance of the topics in the workplace and the value of using playful micro-resources for delivering topics that were deemed less attractive.
...  Lack of skilled workforce (e.g.: in the IT sector).  Lack of equal opportunities in the workplace (McLaughlin et al., 2004, Lane et al., 2015, Gaucher et al., 2011, Quillian et al., 2017.  Different attitude to fulfilment of tasks (Brands-Kehre, 2014). ...
Article
Full-text available
This article discusses the potential of the application of interactive micro-learning content that aims to develop intercultural competences in the workplace, focusing on the pedagogically-informed design of playful interactive learning resources for engaging learners with topics related to cultural risks. Eight topics were identified through our engagement with 154 personnel from multicultural organisations across five European countries, including cross-cultural awareness, stereotypes, and communication. A competence map was then developed, articulating the specific aspects of these topics, which informed online content development. Findings from the evaluation of the online educational materials across the five countries (n=223 participants) suggest the significance and relevance of the topics in the workplace and the value of using playful micro-resources for delivering topics that were deemed less attractive.
... Further along the track research deals with targeted recruitment and perception of job advertisement (Clow et al., 2015;Webster et al., 2020). Particularly in the context of job postings, regarding gender the language used has already been examined in numerous studies (Fernandez & Friedrich, 2011;Gaucher et al., 2011;Hodel et al., 2017;Horvath & Sczesny, 2016). Finally, also at the stage of job interviews, gender and language differences were investigated for a long time (Juodvalkis et al., 2003;Wiley & Eskilson, 1985). ...
Article
Full-text available
Based on previous research about gender differences we investigated whether varying language utilization across gender can be found in standardized text documents such as job applications. To this end, 581 cover letters, CVs and complete application documents were analyzed using linguistic inquiry and word count. Some language differences between men and women could be shown, even if as expected, there were smaller effects than in comparative studies in less formalized contexts. These differences were specific for cover letters and CVs. A differentiated examination showed that gender differences in the cover letters were largely determined by function words. In contrast, differences in the CVs were mainly due to differences in content words. The findings add the context of recruiting to lend support to the framework of gender differences in language across different contexts. Implications for candidates, recruiters and companies are discussed.
... While MBIC dataset itself has labels created by crowdsource workers. We obtained an additional list of most commonly used biases from this source [44] and use it to annotate the dataset further. We went through all the records one by one and added the biased words in the words list. ...
Article
Full-text available
Because of the increasing use of data-centric systems and algorithms in machine learning, the topic of fairness is receiving a lot of attention in the academic and broader literature. This paper introduces Dbias (https://pypi.org/project/Dbias/), an open-source Python package for ensuring fairness in news articles. Dbias can take any text to determine if it is biased. Then, it detects biased words in the text, masks them, and suggests a set of sentences with new words that are bias-free or at least less biased. We conduct extensive experiments to assess the performance of Dbias. To see how well our approach works, we compare it to the existing fairness models. We also test the individual components of Dbias to see how effective they are. The experimental results show that Dbias outperforms all the baselines in terms of accuracy and fairness. We make this package (Dbias) as publicly available for the developers and practitioners to mitigate biases in textual data (such as news articles), as well as to encourage extension of this work.
Article
Since the beginning of this millennium, data in the form of human-generated text in a machine-readable format has become increasingly available to social scientists, presenting a unique window into social life. However, harnessing vast quantities of this highly unstructured data in a systematic way presents a unique combination of analytical and methodological challenges. Luckily, our understanding of how to overcome these challenges has also developed greatly over this same period. In this article, I present a novel typology of the methods social scientists have used to analyze text data at scale in the interest of testing and developing social theory. I describe three “families” of methods: analyses of (1) term frequency, (2) document structure, and (3) semantic similarity. For each family of methods, I discuss their logical and statistical foundations, analytical strengths and weaknesses, as well as prominent variants and applications.
Article
This exploratory study examines the gender parity of public library directors. While there has been considerable research on the role of women library leaders in academic libraries through the 1990s, little is known about the gender parity of public library directors. The membership of the Urban Libraries Council (ULC) serves as the sample size for this study. Gender parity of public library directors reached its zenith in 1999 with 65% of public library directors being women. A little over twenty years later, that number declined to 57.82%, a reduction of about seven percent. This study demonstrates the gender parity of public library directors has not been met.
Article
Public employers struggle with recruiting talents and labor market competition. Research on the understudied topic of employer branding can help address this challenge. This study presents five large-scale, pre-registered field experiments (n = 155,634) aimed at increasing the number of individuals initially interested in a job at a public employer. In social media ads, public sector values served as signaled employer value propositions (EVPs). The results show the importance of target groups and points of difference related to public employers’ organizational type. Significantly fewer women show interest in a job, and for a municipal administration, a fair pay EVP has a negative effect. This study enhances the understanding of potential recruits’ environment- and self-processing, bridges EVPs with public values theory, and provides a missing theoretical link between publicness and recruitment. It shows the importance of testing common assumptions about what works in recruitment in field studies with high external validity (Free postprint available: https://osf.io/azex6/).
Article
Full-text available
The most commonly used method to test an indirect effect is to divide the estimate of the indirect effect by its standard error and compare the resulting z statistic with a critical value from the standard normal distribution. Confidence limits for the indirect effect are also typically based on critical values from the standard normal distribution. This article uses a simulation study to demonstrate that confidence limits are imbalanced because the distribution of the indirect effect is normal only in special cases. Two alternatives for improving the performance of confidence limits for the indirect effect are evaluated: (a) a method based on the distribution of the product of two normal random variables, and (b) resampling methods. In Study 1, confidence limits based on the distribution of the product are more accurate than methods based on an assumed normal distribution but confidence limits are still imbalanced. Study 2 demonstrates that more accurate confidence limits are obtained using resampling methods, with the bias-corrected bootstrap the best method overall.
Article
Full-text available
Recent advances in dyadic data analysis techniques, which treat the dyad, rather than the individual, as the unit of analysis, offer great potential for clinical researchers studying psychopathology. Accordingly, the present article provides readers with a foundation for understanding how the web of interpersonal processes surrounding psychopathology can be modeled and analyzed. The authors start by describing why the analysis of dyadic behaviour may be particularly important for clinical researchers and how issues of dependence that lie at the heart of dyadic data may be productively studied. Next, they describe design issues to consider when studying the interactions of dyads, as well as different kinds of outcome and predictor variables and their data-analytic implications. They introduce the actor-partner interdependence model (APIM), and explain in detail how to estimate it using structural equation modeling (SEM) for both distinguishable and indistinguishable dyads. Extensions of the basic APIM to allow for moderation and mediation, as well as alternative dyadic models involving dyadic latent variables are also covered. Toward the end of the article, the authors describe various approaches for incorporating psychopathology into dyadic SEMs and provide a list of basic questions for clinical researchers to consider when setting up a dyadic model for data analysis.
Article
Stereotype research emphasizes systematic processes over seemingly arbitrary contents, but content also may prove systematic. On the basis of stereotypes' intergroup functions, the stereotype content model hypothesizes that (a) 2 primary dimensions are competence and warmth, (b) frequent mixed clusters combine high warmth with low competence (paternalistic) or high competence with low warmth (envious), and (c) distinct emotions (pity, envy, admiration, contempt) differentiate the 4 competence-warmth combinations. Stereotypically, (d) status predicts high competence, and competition predicts low warmth. Nine varied samples rated gender, ethnicity, race, class, age, and disability out-groups. Contrary to antipathy models, 2 dimensions mattered, and many stereotypes were mixed, either pitying (low competence, high warmth subordinates) or envying (high competence, low warmth competitors). Stereotypically, status predicted competence, and competition predicted low warmth.
Article
Women in the traditionally masculine field of mathematics must contend with stereotypes that allege a sex-based math inability. The threat of being personally reduced to these gender stereotypes can evoke a disruptive state that undermines women's math performance – a situational predicament termed “stereotype threat.” Women are susceptible to stereotype threat whenever they risk fulfilling, or being judged by, a negative gender stereotype that provides a plausible explanation for their behavior in a given domain. This chapter examines the insidious effects that stereotype threat can have on women's performance and aspirations in all quantitative fields. Picture in your mind the typical computer scientist. Now picture the typical librarian. Is the librarian shy and the computer scientist socially awkward? Do they both wear glasses? Are they both inept at sports? Is the computer scientist a male and the librarian a female? Most people can clearly articulate the content of stereotypes targeted at various groups in our society, and gender stereotypes are no exception. As Brown and Josephs (1999) suggest, stereotypes regarding gender differences in math and science ability still pervade contemporary Western thought. In fact, if you assumed that the direction of those gender differences benefited men, you have just displayed personal knowledge of some of the negative stereotypes targeted at women in our culture. This is not meant to imply that you personally endorse those stereotypes – research has shown no relationship between personal beliefs and knowledge of stereotypes (Devine, 1989; Devine & Elliot, 1995).
Book
Part I. From There to Here - Theoretical Background: 1. From visiousness to viciousness: theories of intergroup relations 2. Social dominance theory as a new synthesis Part II. Oppression and its Psycho-Ideological Elements: 3. The psychology of group dominance: social dominance orientation 4. Let's both agree that you're really stupid: the power of consensual ideology Part III. The Circle of Oppression - The Myriad Expressions of Institutional Discrimination: 5. You stay in your part of town and I'll stay in mine: discrimination in the housing and retail markets 6. They're just too lazy to work: discrimination in the labor market 7. They're just mentally and physically unfit: discrimination in education and health care 8. The more of 'them' in prison, the better: institutional terror, social control and the dynamics of the criminal justice system Part IV. Oppression as a Cooperative Game: 9. Social hierarchy and asymmetrical group behavior: social hierarchy and group difference in behavior 10. Sex and power: the intersecting political psychologies of patriarchy and empty-set hierarchy 11. Epilogue.
Chapter
This chapter reviews recent theory and empirical evidence demonstrating the effects of the system justification motive on consequential social and psychological phenomena, as well as the conditions under which these effects are likely to be most pronounced. A review is presented of the theory and evidence demonstrating three conditions that increase the activation of the system justification motive: system threat, perceived system inevitability, and perceptions of personal and system control. A description is made of how, in these conditions, the system justification motive manifests itself in processes of explicit system defense, interpersonal and intergroup perception, and resistance to social change. Throughout, the emphasis is on the contextual nature of these effects, as well as their consequences for the maintenance of social inequality.
Article
In this article, research is reviewed on the emergence of male and female leaders in initially leaderless groups. In these laboratory and field studies, men emerged as leaders to a greater extent than did women. Male leadership was particularly likely in short-term groups and in groups carrying out tasks that did not require complex social interaction. In contrast, women emerged as social leaders slightly more than did men. These and other findings were interpreted in terms of gender role theory, which maintains that societal gender roles influence group behavior. According to this theory, sex differences in emergent leadership are due primarily to role-induced tendencies for men to specialize more than women in behaviors strictly oriented to their group's task and for women to specialize more than men in socially facilitative behaviors.
Article
Recent research has demonstrated that stereotype threat—the concern that others will judge one negatively due to a stereotype that exists about one's group—interferes with women's performance on standardized math and engineering exams. In the current research we find that when a shortened version of the Fundamental of Engineering Exam is described as a test that is diagnostic of ability (i.e., when stereotype threat is high) women perform worse than men on the test. When stereotype threat is reduced, however (by characterizing the test as non-diagnostic or as not producing gender differences), women do just as well as men. The implication of these results for improving the engineering education environment is discussed.
Article
We present evidence that shifting hiring criteria reflects backlash toward agentic (“masterful”) women (Rudman, 1998). Participants (N = 428) evaluated male or female agentic or communal managerial applicants on dimensions of competence, social skills, and hireability. Consistent with past research, agentic women were perceived as highly competent but deficient in social skills, compared with agentic men. New to the present research, social skills predicted hiring decisions more than competence for agentic women; for all other applicants, competence received more weight than social skills. Thus, evaluators shifted the job criteria away from agentic women's strong suit (competence) and toward their perceived deficit (social skills) to justify hiring discrimination. The implications of these findings for women's professional success are discussed.