When Women’s Gains Equal Men’s Losses: Predicting a Zero-Sum Perspective of Gender Status

Abstract

Believing that reduced discrimination against women directly corresponds to increased discrimination against men, referred to as a zero-sum perspective (ZSP), may inhibit further attempts toward gender equality. Based on a sample of 313 men and women, we developed and tested both a general measure and a domain-specific measure of the ZSP of gender status then examined sociodemographics (age, education, political orientation, religious beliefs, and past experience with discrimination) and social dominance orientation as predictors of the ZSP of shifts in gender status. Hostile and modern sexism were examined as potential mediators of this relationship. Structural equation models were computed to examine predictive paths separately for men and women. Although some similarities were found, results showed important differences in predictive paths for women and men, and supported the expected mediating role of sexism in the relationships between sociodemographic predictors and the ZSP. Findings have implications for targeting intervention efforts to enhance a win-win or non-zero-sum perspective that may facilitate efforts toward reducing gender discrimination.

Full text

1 23
Sex Roles
A Journal of Research
ISSN 0360-0025
Sex Roles
DOI 10.1007/s11199-016-0651-9
When Women’s Gains Equal Men’s Losses:
Predicting a Zero-Sum Perspective of
Gender Status
Joelle C.Ruthig, Andre Kehn, Bradlee
W.Gamblin, Karen Vanderzanden &
Kelly Jones

1 23
Your article is protected by copyright and all
rights are held exclusively by Springer Science
+Business Media New York. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

ORIGINAL ARTICLE
When Women’s Gains Equal Men’s Losses: Predicting
a Zero-Sum Perspective of Gender Status
Joelle C. Ruthig
1
&Andre Kehn
1
&Bradlee W. Gamblin
1
&Karen Vanderzanden
1
&
Kelly Jones
1
#Springer Science+Business Media New York 2016
Abstract Believing that reduced discrimination against
women directly corresponds to increased discrimination
against men, referred to as a zero-sum perspective (ZSP),
may inhibit further attempts toward gender equality. Based
on a sample of 313 men and women, we developed and tested
both a general measure and a domain-specific measure of
the ZSP of gender status then examined sociodemographics
(age, education, political orientation, religious beliefs, and past
experience with discrimination) and social dominance orienta-
tion as predictors of the ZSP of shifts in gender status. Hostile
and modern sexism were examined as potential mediators of
this relationship. Structural equation models were computed to
examine predictive paths separately for men and women.
Although some similarities were found, results showed im-
portant differences in predictive paths for women and men,
and supported the expected mediating role of sexism in the
relationships between sociodemographic predictors and the
ZSP. Findings have implications for targeting intervention
efforts to enhance a win-win or non-zero-sum perspective
that may facilitate efforts toward reducing gender
discrimination.
Keywords Gender .Sexism .Discrimination .Social
dominance orientation .Zero-sum
Actual trends in gender discrimination indicate that although
the gender gap is narrowing, it still exists and gender equality
has not yet been reached in the United States (Blau and
Kahn 2007; DeNavas-Walt et al. 2011). An issue receiving
comparably less consideration is how society perceives gen-
der discrimination to have changed over time and how shifts
in discrimination against women are thought to relate to shifts
in discrimination against men. Such research may identify a
source of resistance to further attempts toward gender equal-
ity. If most people believe that gender discrimination is a thing
of the past, motivation to achieve equality for women may be
thwarted. Likewise, the perception that reduced discrimina-
tion against women corresponds directly with increased dis-
crimination against men would presumably impede egalitari-
an efforts, particularly among men. Despite the lack of
supporting evidence that gains in women’srightsleadto
losses for men, recent research found that men indeed viewed
discrimination against men to be increasing concurrently with
decreases in discrimination against women (Bosson et al.
2012;KehnandRuthig2013). Kehn and Ruthig (2013)pos-
ited that this perception is akin to a zero-sum perspective
(ZSP) in which women’s gains (reduced discrimination
against women) are thought to be directly tied to men’slosses
(increased discrimination against men).
The current study builds upon prior gender discrimination
research by directly asking whether individuals endorse a ZSP
regarding shifts in status or discrimination against women and
*Joelle C. Ruthig
joelle.ruthig@und.edu
Andre Kehn
andre.kehn@und.edu
Bradlee W. Gamblin
bradlee.gamblin@my.und.edu
Karen Vanderzanden
karen.vanderzanden@my.und.edu
Kelly Jones
kelly.jones@my.und.edu
1
Department of Psychology, University of North Dakota, 319 Harvard
Street, Stop 8380, Grand Forks, ND 58202, USA
Sex Roles
DOI 10.1007/s11199-016-0651-9
Author's personal copy

men. We also attempted to identify sociodemographic predic-
tors of the ZSP of gender status in order to enable more effec-
tive targeting of initiatives to change this perspective. Finally,
because sociodemographics are largely unalterable, thereby
limiting the potential effectiveness of interventions, we con-
sidered psychosocial mediators of this relationship that would
be more malleable. These objectives are described in detail
following a discussion of key constructs, a theoretical frame-
work, and relevant prior research findings.
Zero-Sum Perspective and Sociodemographics
The Zero-Sum Perspective (ZSP), which originates from game
theory (Nash 1950), states that any measureable gain by one
group is perceived as directly coinciding with equivalent
losses from another group. In other words, the underlying
assumptions of the ZSP are that a finite amount of a given
resource exists and that the gain of that resource by one
group directly corresponds to the loss of that resource by
another group. Within the context of gender discrimination
or status, the ZSP would reflect the belief that any gain
made by women in terms of reduced anti-women bias or
increased status Bcosts^men in terms of increased anti-men
bias or reduced status.
The motivation to endorse a ZSP of gender status can be
explained by a social dominance orientation framework.
According to social dominance orientation, individuals who
belong to the socially dominant group are more likely to
support social-hierarchy-enhancing values and belief sys-
tems whereas individuals belonging to a socially subordi-
nate group are likely to support hierarchy-attenuating
values and belief systems (Pratto et al. 1994; Sidanius
and Pratto 2001). In the case of gender, men have historically
held a socially dominant position with women in a socially
subordinate position. As such, men should bemore likely than
women to endorse the ZSP of gender status, being motivated
by the goal of maintaining their dominant position in the
gender hierarchy.
Holding a zero-sum view of gender status can be detrimen-
tal when directing efforts towards initiatives to enhance equal
opportunity practices in the workplace, for example. Socially-
dominant individuals who endorse a ZSP would be less in-
clined to strive for equality based on the perception that mod-
ifying the current status to benefit others would result in a loss
of resources allocated towards their own group. Hence, it is
critical to determine who these individuals are by identifying
the sociodemographic profile that is indicative of a ZSP of
gender status. In that regard, Bosson et al. (2012)and
Wilkins et al. (2015) found that men endorsed a ZSP of gender
status and discrimination more so than women did. Similarly,
Kehn and Ruthig (2013) found men believed that decreases in
discrimination against women corresponded with increases in
discrimination against men. Conversely, women saw no link
between changes in discrimination against men and women.
Moreover, age differences indicated that compared to their
younger counterparts, men over the age of 35 held the ZSP
of gender discrimination to a greater degree, perhaps because
men in middle-to-late adulthood were raised in a society with
more traditional gender roles compared to younger men (Kehn
and Ruthig 2013).
Within the current study, gender differences in endorse-
ment of the ZSP of gender status were examined, both in terms
of level of endorsement and in predictors of that endorsement.
We tested age, education, political orientation, and strength
of religious beliefs as sociodemographic predictors of the
ZSP of gender status separately for women and men.
Individuals with higher levels of education tend to hold
more egalitarian views than do those with less education
(Fitzpatrick Bettencourt et al. 2011). Likewise a more con-
servative political orientation and strength of one’s reli-
gious beliefs can also influence perspectives on gender
equality (Cokley et al. 2010) and endorsement of the
ZSP (Bosson et al. 2012). Furthermore, because prior per-
sonal experience largely shapes current beliefs, we consid-
ered whether being a past target of gender discrimination
influenced endorsement of the ZSP of gender status. In
addition to these sociodemographic predictors, we assessed
the predictive role of social dominance orientation (Sidanius
and Pratto 2001)ontheZSP.
MediatorsofSociodemographics→ZSP
In addition to identifying sociodemographic predictors of the
ZSP of gender status among men and women, we examined
whether psychosocial factors mediate this link. In particular,
hostile sexism, the explicit endorsement of traditional gender
roles and Bhostility toward women who challenge male
power^(Glick et al. 2004, p. 715), likely corresponds with
the ZSP of gender status. Thus, men who endorse hostile sex-
ismshouldpresumablyalsoholdaZSP.
Whereas some men endorse hostile sexism because it
aims to protect their dominant group status (Sibley et al.
2007), little research has examined why women endorse
hostile sexism. An exception is Becker (2010) who found
that women’s hostile sexism targets norm-deviant women
who contradict traditional role conceptions (e.g., career
women). Thus, some women view efforts to change the
gender power structure as conflicting with Bthe natural
order^and therefore undesirable, even if in their own
interests (Roets et al. 2012).
Modern sexism is another potential mediator of the link
between sociodemographic factors and the ZSP of gender
status. Whereas hostile sexism endorses maintaining women’s
subordination to men, modern sexism reflects a belief that
Sex Roles
Author's personal copy

women are no longer in a subordinate role and that gender
discrimination no longer exists (Becker and Swim 2012;
Swim et al. 1995). This perspective is considered sexism be-
cause it serves to sustain the status quo and diminishes egal-
itarian efforts (Tougas et al. 1995). Presumably, individuals
higher in modern sexism who regard gender discrimination
as obselete would be more likely to hold a ZSP of gender
status than those who are lower in modern sexism. That is,
believing that men and women are already on an equal level,
any further gains benefitting women would presumably be
viewed as Bcosting^men.
In addition to gender, various other sociodemographic fac-
tors (including differences in age, education level, religious
and political beliefs, past experiences with sexist discrimina-
tion, and social dominance orientation) likely contribute to
varying levels of sexist attitudes. Given that younger genera-
tions are less likely than their older counterparts to adhere to
traditional gender roles, age likely corresponds with greater
endorsement of hostile sexism. Support for this assumption
comes from prior research showing a trend towards egalitarian
attitudes among younger generations (Fitzpatrick Bettencourt
et al. 2011). Conversely, because education facilitates more
liberal and egalitarian perspectives on gender (Fitzpatrick
Bettencourt et al. 2011), education level should inversely pre-
dict hostile sexism. Indeed, Glick et al. (2002) found that
higher education levels were associated with less endorsement
of sexist attitudes toward both men and women.Prior research
on the link between religious beliefs and hostile sexism has
been mixed, with some studies showing religious beliefs
predict greater sexism (Tasdemir and Sakalli-Ugurlu 2010),
other research showing religious beliefs to predict less
sexism (Gaunt 2012), and still others showing no link
between religious beliefs and sexism (Maltby et al. 2010).
Also, more conservative political beliefs tend to be more
strongly related to sexism than are more liberal political
beliefs (Fitzpatrick Bettencourt et al. 2011). Past experi-
ences as the target of gender discrimination may also con-
tribute to individuals’levels of hostile and modern sexism
(Becker and Swim 2011). Finally, prior research shows a
positive association between social dominance orientation
and sexism (Sibley et al. 2007).
In addition to endorsement of a broad and overall
ZSP of shifts in gender status, it would be informative
to determine the degree to which the ZSP is endorsed
within specific occupational, economic, political, social,
and familial domains. For example, occupational resources
may be viewed as finite during periods of economic hardship,
potentially enhancing the ZSP of gender status within that
domain. Accordingly, we developed and assessed a multi-
item measure of the ZSP of gender status within multiple
specific domains that have a history of unequal gender status
(occupational, power, economic, political, social status, and
familial) in addition to assessing a broader general ZSP.
Identifying the degree of ZSP endorsement within various
domains will facilitate more effectively targeted intervention
efforts to reduce gender discrimination.
The Present Study
Our objectives in the present study were to examine gender
differences in endorsement of the ZSP of gender status and to
identify sociodemographic predictors of the ZSP, as well as to
determine whether hostile and modern sexism mediate those
links. Our study builds upon prior research examining the
corresponding shifts in perceived discrimination against men
and women (Kehn and Ruthig 2013) by directly asking indi-
viduals whether they endorse a ZSP regarding changes in
gender status, both in general and within multiple specific
domains. We expected that men would endorse the ZSP of
gender status to a greater extent than would women. We also
expected that age, stronger religious beliefs, and a more
conservative political orientation would predict greater en-
dorsement of the ZSP, whereas education level would pre-
dict less endorsement of the ZSP. Also, individuals who are
higher in social dominance orientation should endorse the
ZSP to a greater degree than should those who are lower in
social dominance orientation. Indeed, recent research pro-
vides evidence of the positive association between social
dominance orientation and the ZSP (Wilkins et al. 2015). We
expanded upon this prior research by examining the predictive
role of social dominance orientation on the ZSP beyond the
effects of multiple sociodemographic factors and both hostile
and modern sexism.
Finally, we explored whether being a past target of gender
discrimination predicted endorsement of the ZSP. For men,
being the target of gender discrimination was expected to pre-
dict endorsement of the ZSP such that reflecting on past, per-
sonal experiences of anti-men bias would contribute to the
win-lose view of any gains favoring women. We also exam-
ined whether this past discrimination →ZSP link was similar
or different among women as compared to men. Sexist beliefs
were expected to mediate these relationships in the following
manner: age, strength of religious beliefs, a more conservative
political view, past personal experience with discrimination,
less education, and a stronger social dominance orientation
should predict greater hostile and modern sexism, which in
turn, should predict greater endorsement of the ZSP.
To summarize, we address three main research questions
with the current study. (a) Do sociodemographic factors of
age, education, political orientation, strength of religious
beliefs, and past personal discrimination predict the ZSP
of gender status? (b) Do hostile and modern sexism
mediate the associations between sociodemographic predic-
tors and the ZSP? (c) Do these predictive relationships differ
for women and men?
Sex Roles
Author's personal copy

Method
Participants and Procedure
We initially recruited 387 adult men and women, who
represented all four regions of the United States, by using
Amazon’s Mechanical Turk—an online data collection tool
that yields reliable, demographically diverse samples
(Buhrmester et al. 2011). Participants were compensated
$0.25 for completing measures of sociodemographics, social
dominance orientation, hostile and modern sexism, and the
ZSP of gender status. Due to random missing data, 15
(3.9 %) participants’data were excluded from subsequent
analyses. As detailed in the following measures section, data
from 59 participants were also omitted from analyses,
resulting in a final sample of 313 participants (142 men and
171 women) on which all subsequent analyses are based.
On average, participants were 35.12 years old (SD =12.11)
and spanned young through older adulthood (range = 18–
74 years-old). The vast majority of participants were Caucasian
(n= 285, 91 %), with the remaining 9 % of participants
identifying as African American (n=11,3.5%),Asian
(n=9,2.9%),Hispanic(n= 4, 1.3 %), Native American
(n= 1, .3 %), or Other (n=3,1%).Most participants
(n= 216, 69 %) had at least some college education, whereas
96 (30.7 %) participants reported having at least completed
high school/high school equivalency, and one participant
reported having less than a high school education.
Measures
Sociodemographic Measures
In addition to reporting their age, race, and highest level of
education attained, participants were also asked to indicate the
strength of their religious beliefs, on a scale from 1 (not at all
strong)to7(very strong), and BWhat is your political
orientation,^with response options ranging from 1 (liberal)
through 7 (conservative). A final sociodemographic measure
assessed participants’personal experience of being the target
of gender discrimination: BTo what extent have you personally
been discriminated against because of your gender?^,ratedon
a scale from 1 (not at all)to7(very much so).
Social Dominance Orientation
Sidanius and Pratto’s(2001) 16-item scale was used to assess
participants’social dominance orientation (reported reliabilities
range from α= .89 to .92; Shook et al. 2016). Participants
indicated the degree to which they agreed with statements such
as BSome groups of people are just more worthy than others,^
with response options ranging from 1 (Strongly disagree)
through 7 (Strongly agree). Half the items were reverse coded
andtheneachparticipant’s responses to all items were averaged
to create a composite score in which higher scores indicate
stronger social dominance orientation (α= .94).
Hostile Sexism
Participants completed the 11-item Hostile Sexism subscale of
the Ambivalent Sexism Inventory (Glick and Fiske 1996;re-
ported alphas across six samples ranged from .80–.92. An
example is: BWomen seek to gain power by getting control
over men,^with response options ranging from 0 (disagree
strongly)through5(agree strongly). Responses to three items
(e.g.: BFeminists are making entirely reasonable demands
of men^) were reverse coded and then each participant’s
responses to all items were averaged to create a composite
score in which higher scores indicate greater hostile sexism
(α=.91).
Modern Sexism
Swim et al.’s(1995) 8-item Modern Sexism scale was used to
assess participants’modern sexist beliefs (α= .84 as reported
by Swim et al. 1995). For example: BDiscrimination against
women is no longer a problem in the United States,^with
response options from 1 (strongly disagree)to5(strongly
agree). Responses to three items were reversed and then each
participant’s responses to all items were averaged to create
a composite score in which higher scores indicate greater
modern sexism (α=.87).
Zero-Sum Perspectives of Gender Status
Two separate measures were used to assess the ZSP of gender
status. First, as a broad general ZSP measure, participants
indicated the extent to which they believed that Bdeclines in
discrimination against women are directly related to increased
discrimination against men,^with response options ranging
from 1 (not at all) through 10 (very much). Participants also
had the option of selecting BNot applicable: discrimination
against women has not been reduced.^
In addition to this general measure, participants responded
to six additional items that assessed the ZSP of gender status
with specific domains: occupational (BMore good jobs for
women mean fewer good jobs for men^), power (BThe more
power women gain, the less power men have^), economic
(BWomen’s economic gains translate into men’s economic
losses^), political (BThe more influence women have in
politics, the less influence men have in politics), social
status (BAs women gain more social status, men lose social
status), and familial (BMore family-related decision making
for women means less family-related decision making for
men^). Responses options for each item ranged from 1
(not at all) through 10 (very much) and participants were
Sex Roles
Author's personal copy

provided with a BNot applicable^response option. Data
from 59 participants who selected the not applicable option
for one or more of the ZSP items were omitted from analyses.
In comparison to the remaining sample, those omitted partici-
pants indicated weaker religious beliefs (M=2.87,SD = 2.25
vs. M=3.59,SD = 2.24), t(370) = 2.31, p=.02,andlower
social dominance orientation (M=1.93,SD = .98 vs. M= 2.48
SD =1.22),t(370) = 3.39, p=.01.
Responses for all six items from remaining participants
were averaged to create the composite measure in whichhigher
scores reflect greater endorsement of the ZSP (α= .93) An
exploratory factor analysis (EFA) with principle axis factoring
and direct oblimin rotation revealed strong evidence of unidi-
mensionality, with a one-factor solution explaining 70.6 % of
the variance for the domain-specific ZSP measure.
Results
Preliminary Analyses
Prior to addressing the main research questions, descriptive
statistics and bivariate correlations were computed to examine
associations among study variables, and a measurement
model was computed to assess item-to-factor loadings.
As reported in Table 1, means, standard deviations, and
bivariate correlations among study variables were computed sep-
arately for men and women. As expected, men (M
GZSP
=4.39,
SD =2.59;M
DZSP
=4.00,SD = 2.29) endorsed the ZSP of
gender status to a significantly greater degree than did women
(M
GZSP
= 3.32, SD = 2.27; M
DZSP
= 2.87, SD = 1.91),
t(311) = 3.90, p< .001 and t(311) = 4.73, p< .001, respectively.
Many correlations were similar for men and women: A
more conservative political view was associated with stronger
religious beliefs, greater social dominance orientation, higher
levels of both hostile and modern sexism, and greater endorse-
ment of the ZSP (see Table 1). Likewise, stronger religious
beliefs were associated with greater social dominance orien-
tation and modern sexism, as well as greater endorsement of
domain-specific ZSP. Higher levels of education were associ-
ated with less hostile sexism. Finally, greater social domi-
nance orientation was associated with higher levels of both
types of sexism (which were positively correlated with each
other) and greater endorsement of both measures of ZSP
(which were positively correlated with each other). In turn,
both types of sexism were associated with greater endorse-
ment of both ZSP measures.
Despite these gender similarities, several other bivariate
correlations differed for men and women. Being older was
associated with a more conservative political orientation for
women but not for men. Conversely, older men reported less
experience with past personal discrimination but this relation-
ship did not occur for women. Relationships with education
also differed by gender. For women but not for men, higher
education was associated with a less conservative political
orientation, less social dominance orientation, and less
modern sexism. Among women but not among men,
stronger religious beliefs were associated with greater hos-
tile sexism. Finally, more past personal discrimination was
associated with less modern sexism for women but not for
men. Conversely, greater past personal discrimination was
associated with stronger endorsement of the general ZSP
for men but not for women. Together, with the significant
mean differences in ZSP, these gender differences in cor-
relations provide support for examining predictors of the
Tabl e 1 Means, standard deviations, and bivariate correlations for all study variables for men and women
Vari a b l e s Me n M(SD) Women M(SD)12 34567 8 910
1. Age 35.13 (11.54) 35.11 (12.60) –.07 .19* .12 .01 −.04 .05 .02 −.06 .12
2. Educ 3.35 (1.07) 3.33 (1.11) .18* –−.19* −.14 .09 −.16* −.27*** −.22** −.11 −.08
3. Polit 3.70 (1.75) 3.47 (1.63) .07 −.04 –.55*** −.13 .36*** .47*** .38*** .27*** .31***
4. Relig 3.47 (2.21) 3.78 (2.26) .08 −.01 .56*** –.05 .15* .31*** .20* .14 .17*
5. PrsDisc 1.94 (1.10) 2.88 (1.40) −.22** −.14 −.01 .04 –−.07 −.06 −.27*** .06 .07
6. SDO 2.84 (1.25) 2.12 (1.06) −.06 −.01 .42** .22** .08 –.54*** .45*** .41*** .53***
7. HS 2.46 (1.07) 1.77 (1.17) −.06 −.25** .31*** .14 .13 .45*** –.59*** .50*** .47***
8. MS 2.80 (.77) 2.31 (.77) .13 −.04 .45** .27** .10 .49*** .53*** –.38*** .36***
9. GZSP 4.39 (2.59) 3.32 (2.27) −.05 −.21* .18* .16 .34*** .25** .48*** .36*** –.47***
10.DZSP 4.00 (2.29) 2.87 (1.91) .08 −.05 .22** .17* .12 .33*** .53*** .26** .48*** –
Educ Education level, Polit political orientation where higher scores indicate a more conservative orientation and lower scores indicate a more liberal
orientation, Relig strength of religious beliefs, PrsDisc past personal discrimination, SDO social dominance orientation, HS hostile sexism, MS modern
sexism, GZSP general zero sum perspective, DZSP domain-specific zero sum perspective. Correlations for women are reported above the diagonal of the
correlation matrix; Correlation for men are reported below the diagonal
*
p< .05.
**
p< .01.
***
p<.001
Sex Roles
Author's personal copy

ZSP separately for men and women within separate struc-
tural equation models. Prior to describing those analyses,
we conducted preliminary tests of our hostile sexism,
modern sexism, and ZSP scales.
To maximize the fit of the measurement model and more
accurately examine the relationships between latent variables
in the model (Little et al. 2002), we created parcels of two
items each for the Modern Sexism and Hostile Sexism scales.
These parcels were created by pairing items together that
shared thematic content (as recommended by Little et al.
2002). Parcels for social dominance orientation were also cre-
ated and consisted of four items each, with parcels being based
on thematic content of the items. Inter-item reliability was
strong for all three scales when using parcels (α
MS
= .82,
α
HS
=.90,α
SDO
= .96); in addition, the parcels led to stronger
unidimensionality for the scales, with one factor solutions
explaining between 55.08 % (for modern sexism) and
84.11 % of the item variance. Subsequent data analysis was
conducted using these parcels. The newly developed indi-
vidual items for the domain-specific ZSP scale were not
parceled because we wanted to examine their individual
loadings within the measurement model.
Next, a CFA was conducted on hostile sexism, modern
sexism, domain-specific ZSP, and social dominance orienta-
tion to further ensure unidimensionality and to establish con-
struct, convergent, and discriminant validity within and be-
tween the four latent constructs. Model fit for the CFA (and
for the subsequent SEM analyses) was assessed using four
goodness-of-fit indices: χ
2
, standardized root-mean-square re-
sidual (SRMR), root-mean-square error of approximation
(RMSEA), and comparative fit index (CFI). As recommended
(Schreiber et al. 2006), we used fit cutoffs of ≤.08 for SRMR,
≤.06 for RMSEA, and ≥.95 for CFI.
The CFA model indicated adequate fit, χ
2
(146) = 413.74,
p< .01 (RMSEA = .08, CFI = .95, SRMR = .06). Furthermore,
the model suggested both strong convergent validity (Average
variance extracted: AVE
MS
=.74,AVE
HS
= .83,
AV E
ZSP
=.84,AVE
SDO
= .92) and discriminant validity
(all average AVEs higher than their respective r
2
).
Therefore, the measurement model was accepted and struc-
tural models were tested using these latent variables.
Main Analyses
To address the first main research question of which
sociodemographic factors predict the ZSP of gender status
and the second main research question of whether those pre-
dictive effects are mediated by hostile and modern sexism,
separate structural equation models (using AMOS 22.0) were
computed for men and for women. Within the model, age,
education, strength of religious beliefs, political orientation,
past personal discrimination, and social dominance orienta-
tion were included as exogenous variables; hostile sexism
and modern sexism as mediator variables; and the general
and domain-specific ZSP measures as outcome variables.
These SEMs also enabled us to address the third research
question, namely to identify whether these predictive relation-
ships differed between men and women.
Predicting Men’s Endorsement of the ZSP
For men, model estimation including all direct and indirect
paths revealed good fit overall, χ
2
(244) = 448.83, p<.01
(SRMR = .07, RMSEA = .08, CFI = .91). The model ex-
plained 38 % of the variance in hostile sexism, 37 % of the
variance in modern sexism, 33 % of the variance for the gen-
eral ZSP measure, and 43 % for the domain-specific ZSP
measure. We then computed a restricted SEM model for
men that only included paths that were significant in the initial
model. The overall fit of the restricted model was very similar
to the initial model, χ
2
(263) = 469.77, p< .01 (SRMR = .08,
RMSEA = .08, CFI = .91). Likewise, a direct comparison
indicated that the restricted model did not significantly differ
from the original model, Δχ
2
(19) = 20.94, p=.34.The
restricted model explained 35 % of the variance in hostile
sexism, 34 % of the variance in modern sexism, 31 % of
the variance for the general ZSP measure, and 37 % for
the domain-specific ZSP measure.
The restricted structural model for men revealed only one
significant direct path predicting the ZSP (see Fig. 1). That is,
personal experience as the target of past discrimination pre-
dicted greater endorsement of the general ZSP measure; con-
versely, there was no significant direct or indirect path be-
tween past discrimination and the domain-specific measure
of ZSP. The model for men also yielded several significant
indirect paths in terms of sociodemographic measures
predicting sexism, which in turn, predicted the ZSP. Less ed-
ucation and greater social dominance orientation predicted
more hostile sexism, and greater hostile sexism predicted
stronger endorsement of both general ZSP and domain-
specific ZSP. A more conservative political view and greater
social dominance orientation predicted higher levels of mod-
ern sexism but modern sexism did not significantly predict
endorsement of either domain-specific or general ZSP. Thus
no indirect effects were observed through modern sexism.
Predicting Women’s Endorsement of the ZSP
For women, the initial model testing all possible direct and
indirect paths indicated good fit overall, χ
2
(244) = 465.49,
p< .01 (SRMR = .06, RMSEA = .07, CFI =. 93). The model
explained 45 % of the variance in hostile sexism, 32 % of the
variance in modern sexism, 33 % for the general ZSP mea-
sure, and 44 % of the variance for the domain-specific mea-
sure. We then recomputed a restricted SEM model for women
that only included paths that were significant in the initial
Sex Roles
Author's personal copy

model. The overall fit of the restricted model was similar to the
initial model, χ
2
(261) = 483.08, p<.01(SRMR=.07,
RMSEA = .07, CFI = .93). A direct comparison indicated that
the restricted model did not significantly differ from the original
model, Δχ
2
(17) = 17.59, p= .42. The restricted model ex-
plained 45 % of the variance in hostile sexism, 32 % of the
variance in modern sexism, 30 % for the general ZSP measure,
and 40 % of the variance for the domain-specific measure.
The restricted structural model for women showed many
similarities to the men’s structural model (see Fig. 2). As with
men, greater social dominance orientation predicted higher
levels of hostile sexism, which subsequently predicted en-
dorsement of both general ZSP and domain-specific ZSP. In
addition, more conservative political beliefs and social domi-
nance orientation predicted modern sexism for women, and
modern sexism failed to predict either domain-specific or gen-
eral ZSP, echoing the paths found for men.
Apart from these similarities, the structural model for wom-
en showed many unique relationships in comparison to the
model for men (see Fig. 2). Specifically, no sociodemographic
variables directly predicted domain-specific ZSP for men,
but for women, past experience as a target of discrimina-
tion and more social dominance orientation predicted
stronger endorsement of domain-specific ZSP. Many indirect
relationships, partially mediated by sexism also differed by
gender. Whereas political orientation was not significantly
related to hostile sexism for men, more conservative political
beliefs predicted greater hostile sexism for women. Although
less education predicted greater hostile sexism for men, this
association was not found among women. Also unlike men,
women’s modern sexism was predicted by having less expe-
rience as a target of discrimination.
Discussion
A potential barrier to gender equality is the Zero-Sum
Perspective (ZSP) of gender status: the belief that any gains
toward equality made by women Bcost^men. The current
study determined the degree to which men and women en-
dorse both a general overall and a domain-specific ZSP of
gender status as well as identified sociodemographic predic-
tors of that endorsement and sexist beliefs that partially medi-
ate the predictive relationship.
Endorsing the Zero-Sum Perspective
Preliminary findings indicated strong internal reliability of the
multi-item domain-specific ZSP measure. This suggests that
individuals who have a win-lose perspective of shifts in gen-
der status in the workplace (i.e., women’s employment gains
mean men’s employment losses) are likely to have that same
perspective of shifts in gender status within political, familial,
economic, and social domains. Thus, the ZSP of gender status
appears to be pervasive, and the degree to which this compet-
itive, win-lose perspective provides resistance to gender
equality efforts in one domain likely reflects a similar barrier
to gender equality in multiple other domains. Moreover, the
domain-specific measure of ZSP was significantly associated
with the general ZSP measure, indicating that a broadly held
win-lose perspective of shifts in gender status likely reflects
one’s view within specific employment or social domains.
Likewise, the win-lose perspective within a specific employ-
ment or social domain would likely be consistent with one’s
overall perception of gender status and discrimination in gen-
eral. As such, intervention efforts to replace the ZSP with a
.44**
.23**
-.19*
Educat
Religio
Belie
Politic
Orienta
Person
Discrimin
Social Dom
Orienta
.56**
Age
tion
ous
efs
cal
ation
nal
nation
minance
ation
-.26**
.25**
.54**
.41**
e
*
.41**
*
e1
e2
Hostile S
Modern
.27*
Sexism
Sexism
**
.48**
.67**
General ZS
Gender Sta
Domain-specific Z
Gender Statu
*
P of
atus
ZSP of
us
e3
e4
.29**
Fig. 1 Restricted SEM predicting men’s endorsement of the zero-sum perspective of gender status. *p< .05. **p≤.01
Sex Roles
Author's personal copy

less competitive view of gender may be more effective when
based on a general approach rather than focusing on one spe-
cific life domain.
Along with its high internal reliability, the measurement
model demonstrated both convergent and discriminant valid-
ity of the multi-domain measure of the ZSP of gender status.
Together, these psychometric results indicate that this is an
effective measure to assess how individuals view changes in
gender status. Future research should continue to examine its
utility in assessing this Bwin-lose^perspective of gender sta-
tus. In particular, assessing the ZSP measure among broader
samples and within experimental designs would provide in-
sight into the limits of the measure’s generalizability and the
degree to which this multi-domain view of shifts in gender
status can be directly altered.
In an attempt to determine who endorses the ZSP of gender
status, the current study identified multiple sociodemographic
predictors of this perspective. Although prior research exam-
ining perceived shifts in gender discrimination over time
found that only men held the ZSP (Bosson et al. 2012; Kehn
and Ruthig 2013), the current findings indicate that both men
and women endorsed the ZSP, although men did so to a great-
er degree. Thus, some women view anti-discrimination or
gains made by women as directly coinciding with increased
discrimination against or losses for men. This finding high-
lights a need to target both genders when attempting to foster a
cooperative view of reducing gender discrimination to replace
the competitive Bwin-lose^ZSP.
In addition to identifying the need to focus on both men and
women in efforts to reduce the ZSP, the current findings indi-
cate that past personal experience as a target of gender dis-
crimination directly predicted endorsement of the ZSP for
both women and men. Having been discriminated against be-
cause of one’s gender is associated with a greater likelihood
that those individuals will view status gains or reducing bias
toward men and women as a competition. This suggests that a
person’s perception of being the target of past discrimination
may be a hindrance, rather than an asset, to gender equality
efforts. Instead of empathy or personal investment in reducing
discrimination similar to what one has personally suffered,
such prior experiences may stand in the way of equality ef-
forts. This association may be due to believing that past per-
sonal gender discrimination was a direct result of the other
gender being favored in some way. In this sense, such indi-
viduals may perceive their past discrimination experiences as
direct evidence of the zero-sum perspective or win-lose ap-
proach to shifts in gender status.
Despite both women and men endorsing the zero-sum per-
spective and that endorsement being predicted by past person-
al discrimination, there were important gender differences in
level of ZSP endorsement and predictors of that endorsement.
Notably, social dominance orientation directly predicted
domain-specific endorsement of the ZSP for women but not
for men. Instead, the role of social dominance orientation on
men’s endorsement of the ZSP was partially mediated by sex-
ism, which played somewhat different mediating roles for
men and women as subsequently described.
The Mediating Role of Sexism
Although past personal discrimination had direct predictive as-
sociations with the ZSP, the remaining sociodemographic pre-
dictors either had non-significant (i.e., age and strength of reli-
gious beliefs) or indirect predictive effects, partially mediated
Age
Education
Religious
Beliefs
Political
Orientation
Personal
Discrimination
Domain-specific ZSP
of Gender Status
Hostile Sexism
Modern Sexism
Social Dominance
Orientation
.28**
.54**
.37**
.30**
.24**
-.25** .13*
.49** .37** .33**
.15**
.53**
.16* .37**
General ZSP of
Gender Status
.48**
e1
e2
e3
e4
Fig. 2 Restricted SEM predicting women’s endorsement of the zero-sum perspective of gender status. *p<.05.**p≤.01
Sex Roles
Author's personal copy

by hostile sexism (i.e., education, political orientation, and for
men, social dominance orientation). Perhaps this is Bgood
news^for gender equality efforts, given that hostile sexism is
relatively more malleable than sociodemographic factors such
as age or strength of one’s religious beliefs. Moreover, hostile
sexism was a consistent predictor of both overall and domain-
specific ZSP for both men and women. Our findings also
showed that regardless of gender, individuals with a stronger
social dominance orientation tended to have the highest
levels of hostile sexism. In addition, women who had a
more conservative political orientation and men with less
education also had higher levels of hostile sexism. Thus,
these may be key characteristics to focus on when targeting
efforts to reduce hostile sexism in an attempt to minimize
the ZSP of gender status.
Practice Implications
Some recent research has shown success in reducing hostile
sexism in both men and women via experiential and learning-
based intervention strategies (Zawadzki et al. 2014). Other
research suggests that although focusing on hostile sexism
may be an effective approach for interventions in terms
of alleviating the view of women’s gains towards equality
in a negative manner, such initiatives may differ by gen-
der. Specifically, women high in hostile sexism tend to
identify with traditional female subtypes (e.g., housewife)
and distance themselves from non-traditional subtypes
(e.g., career woman), engaging in sexism against the latter
group (Becker 2010). As such, intervention efforts that
narrow the gap and promote common bonds between tra-
ditional and non-traditional women may help some women
to identify with, and become supportive of, non-traditional
women.
For men, effective interventions to reduce the ZSP via di-
minished sexist beliefs may entail a social norms approach
clarifying that most men overestimate the degree of their
peer’s sexism towards women. Such information pressures
men to conform to the social norm by reducing their own
sexist attitudes (Kilmartin et al. 2008). Likewise, education
regarding the harmful effects of sexism poses another possi-
bility for reducing the ZSP among men via decreasing their
level of modern sexism (Becker and Swim 2012).
Whereas hostile sexism was found to partially mediate the
links between various sociodemographics and both the gener-
al and domain-specific ZSP for men and women, modern
sexism was not a mediator in this regard. Instead, modern
sexism did not significantly predict ZSP among women or
men. We had reasoned that individuals high in modern sexism
would be more likely to hold a ZSP of gender status because
believing that men and women are already on an equal level,
any further status gain for women would presumably be
viewed as Bcosting^men in terms of a decrease in their status.
However, any association between modern sexism and ZSP
appears to be secondary to the strong, consistent link between
hostile sexism and ZSP for both men and women.
Finally, the current results indicate that minimizing social
dominance orientation may also contribute to decreasing both
hostile and modern sexism for men and women, as well as less
endorsement of the ZSP, particularly among women. Some
recent initiatives suggest positive inter-group contact may de-
crease social dominance orientation within interracial contexts
(Dhont et al. 2014; Shook et al. 2016). Subsequent research
efforts could be directed toward developing similar interven-
tion strategies of positive contact experiences between gen-
ders in attempt to reduce social dominance orientation, and in
turn, decrease endorsement of the ZSP of gender status.
Limitations
Despite the aforementioned implications, the current findings
are not without some limitations. In particular, all measures
were self-report and thus subject to socially desirable
responding biases. For example, participants may have per-
ceived that strongly endorsing hostile sexism statements
would go against what is deemed to be socially acceptable.
Although we attempted to minimize social desirability effects
by assuring participants of their anonymity and having them
complete the survey privately, lower scores on measures such
as hostile and modern sexism may reflect some degree of
social desirability. Another limitation is that we did not assess
engagement in actual gender discrimination. The current find-
ings did establish links between sexism and endorsement of
the ZSP, however, it remains unclear whether endorsing this
perspective directly contributes to gender discriminatory be-
havior. Subsequent research including a gender discrimination
outcome would enable assessment of the association between
endorsing the zero-sum perspective of gender status and en-
gaging in acts of gender discrimination. Finally, the general-
izability of our findings is limited to adults residing within the
United States so that subsequent research should attempt to
examine predictors of the zero-sum perspective across other
countries and within non-Western cultures.
Conclusion
Although narrowing, the gender gap continues to exist
in the United States. In an attempt to identify a potential
source of resistance to further gains toward gender equality,
our study focused on how shifts in women’s status are thought
to relate to shifts in men’s status. The current findings contrib-
ute to existing gender discrimination research in terms of
highlighting a broad win/lose perspective held by both men
and women that may hinder gender equality efforts in multiple
domains, as well as identifying sociodemographic and mallea-
ble psychosocial predictors of that perspective.
Sex Roles
Author's personal copy

References
Becker, J. C. (2010). Why do women endorse hostile and benevolent
sexism? The role of salient female subtypes and internalization of
sexist contents. Sex Roles, 62,453–467. doi:10.1007/s11199-009-
9707-4.
Becker, J. C., & Swim, J. (2011). Seeing the unseen attention to daily
encounters with sexism as a way to reduce sexist beliefs. Psychology
of Women Quarterly, 35, 227–242. doi:10.1177/0361684310397509.
Becker, J. C., & Swim, J. (2012). Reducing endorsement of benevolent
and modern sexist beliefs: Differential effects of addressing harm
versus pervasiveness of benevolent sexism. Social Psychology, 43,
127–137. doi:10.1027/1864-9335/a000091.
Blau, F. D., & Kahn, L. M. (2007). The gender pay gap: Have women
gone as far as they can? Academy of Management Perspectives, 21,
7–23. doi:10.5465/AMP.2007.24286161.
Bosson, J. K., Vandello, J. A., Michniewicz, K. S., & Lenes, J. G. (2012).
American men’sandwomen’s beliefs about gender discrimination:
For men, it’s not quite a zero-sum game. Masculinities and Social
Change, 1,210–239. doi:10.4471/mcs.2012.14.
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's mechanical
Turk: A new source of inexpensive, yet high-quality, data? Perspectives
on Psychological Science, 6,3–5. doi:10.1177/1745691610393980.
Cokley, K. O., Tran, K., Hall-Clark, B., Chapman, C., Bessa, L., Finley,
A., et al. (2010). Predicting student attitudes about racial diversity
and gender equity. Journal of Diversity in Higher Education, 3,
187–199. doi:10.1037/a0020467.
DeNavas-Walt,C.,Proctor,B.D.,&Smith,J.C.(2011).Income, poverty,
and health insurance coverage in the United States: 2011 (U.S. Census
Bureau, Current Population Reports, P60–226). Washington, DC:
U.S. Government Printing Office.
Dhont, K., Van Hiel, A., & Hewstone, M. (2014). Changing the ideolog-
ical roots of prejudice: Longitudinal effects of ethnic intergroup
contact on social dominance orientation. Group Processes and
Intergroup Relations, 17,27–44. doi:10.1177/1368430213497064.
Fitzpatrick Bettencourt, K. E., Vacha-Haase, T., & Byrne, Z. S. (2011).
Older and younger adults’attitudes toward feminism: The influence
of religiosity, political orientation, gender, education, and family.
Sex Roles, 64,863–874. doi:10.1007/s11199-011-9946-z.
Gaunt, R. (2012). BBlessed is he who has not made me a women^:
Ambivalent sexism and Jewish religiosity. Sex Roles, 67,477–487.
doi:10.1007/s11199-012-0185-8.
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., Lameiras, M., & Castro, Y. R. (2002). Education and
Catholic religiosity as predictors of hostile and benevolent
sexism toward women and men. Sex Roles, 47, 433–441.
doi:10.1023/A:1021696209949.
Glick, P., Lameiras, M., Fiske, S. T., Eckes, T., Masser, B., Volpato, C.,
et al. (2004). Bad but bold: Ambivalent attitudes toward men predict
gender inequality in 16 nations. Journal of Personality and Social
Psychology, 86,713–728. doi:10.1037/0022-3514.86.5.713.
Kehn, A., & Ruthig, J. C. (2013). Perceptions of gender discrimination
across six decades: The moderating roles of gender and age. Sex
Roles, 69,289–296. doi:10.1007/s11199-013-0303-2.
Kilmartin, C., Smith, T., Green, A., Heinzen, H., Kuchler, M., &
Kolar, D. (2008). A real time social norms intervention to
reducemalesexism.Sex Roles, 59, 264–273. doi:10.1007
/s11199-008-9446-y.
Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K.
(2002). To parcel or not to parcel: Exploring the question,
weighing the merits. Structural Equation Modeling, 9,151–
173. doi:10.1207/S15328007SEM0902_1.
Maltby, E. L., Hall, M. E. L., Anderson, T. L., & Edwards, K. (2010).
Religion and sexism: The moderating role of participant gender. Sex
Roles, 62,615–622. doi:10.1007/s11199-010-9754-x.
Nash, J. F. (1950). The bargaining problem. Econometrica, 18,155–162.
doi:10.2307/1907266.
Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social
dominance orientation: A personality variable predicting social and
political attitudes. Journal of Personality and Social Psychology, 67,
741–763. doi:10.1037/0022-3514.67.4.741.
Roets, A., Van Hiel, A., & Dhont, K. (2012). Is sexism a gender issue? A
motivated social cognition perspective on men’s and women’s sexist
attitudes toward own and other gender. European Journal of
Personality, 26,3
50–359. doi:10.1002/per.843.
Schreiber, J. B., Nora, A., Stage, F. K., Barlow, E. A., & King, J. (2006).
Reporting structural equation modeling and confirmatory factor
analysis results: A review. The Journal of Educational Research,
99,323–338. doi:10.3200/JOER.99.6.323-338.
Shook, N. J., Hopkins, P. D., & Koech, J. M. (2016). The effect of
intergroup contact on secondary group attitudes and social domi-
nance orientation. Group Processes & Intergroup relations, 19,
328–342. doi:10.1177/1368430215572266.
Sibley, C. G., Wilson, M. S., & Duckitt, J. (2007). Antecedents of men’s
hostile and benevolent sexism: The dual roles of social dominance
orientation and right-wing authoritarianism. Personality and Social
Psychology Bulletin, 33, 160–172. doi:10.1177/0146167206294745.
Sidanius, J., & Pratto, F. (2001). Social dominance: An intergroup theory
of social hierarchy and oppression. Cambridge: Cambridge
University Press.
Swim, J. K., Aikin, K. J., Hall, W. S., & Hunter, B. A. (1995).
Sexism and racism: Old-fashioned and modern prejudices.
Journal of Personality and Social Psychology, 68, 199–214.
doi:10.1037/0022-3514.68.2.199.
Tasdemir, N., & Sakalli-Ugurlu, N. (2010). The relationship between
religiosity and ambivalent sexism among Turkish students. Sex
Roles, 62,420–426. doi:10.1007/s11199-009-9693-6.
Tougas, F., Brown, R., Beaton, A. M., & Joly, S. (1995). Neosexism: Plus
a change, plus c’est pareil. Personality and Social Psychology
Bulletin, 21,842–849. doi:10.1177/0146167295218007.
Wilkins, C. L., Wellman, J. D., Babbitt, L. G., Toosi, N. R., & Schad, K.
D. (2015). You can win but I can’t lose: Bias against high-status
groups increases their zero-sum beliefs about discrimination.
JournalofExperimentalSocialPsychology,15,1–14. doi:10.1016
/j.jesp.2014.10.008.
Zawadzki, M. J., Shields, S. A., Danube, C. L., & Swim, J. K.
(2014). Reducing the endorsement of sexism using experiential
learning: The workshop activity for gender equity simulation
(WAGES). Psychology of Women Quarterly, 38,75–92.
doi:10.1177/0361684313498573.
Sex Roles
Author's personal copy