Content uploaded by Peter Glick
Author content
All content in this area was uploaded by Peter Glick on Jun 17, 2014
Content may be subject to copyright.
It’s For Your Own Good - 1
It’s for Your Own Good:
Benevolent Sexism and Women’s Reactions to Protectively Justified Restrictions
Miguel Moya Peter Glick
Universidad de Granada Lawrence University
Francisca Expósito Soledad De Lemus
Universidad de Granada Universidad de Granada
Joshua Hart
Lawrence University
WORD COUNT: 9,983
Keywords: Paternalism, Discrimination, Benevolent Sexism, Close Relationships
Corresponding author:
Peter Glick
Department of Psychology
Lawrence University
PO Box 599
Appleton, WI 54912-0599
Email: glickp@lawrence.edu
Phone: (920) 832-6707
Fax: (920) 832-6962
It’s For Your Own Good - 2
Abstract
Three studies investigated women’s reactions to ostensibly protective restrictions. In Study
1, only benevolently sexist women accepted a protectively-justified (hypothetical)
prohibition against driving on a long trip, but only when imposed by a husband (not a co-
worker). In Study 2, when women’s actual romantic partners opposed their participation in
a practicum counseling dangerous men, most reacted positively to a personalized
protective justification (“I am concerned for your safety”), but only benevolently sexist
women reacted positively when no justification was given. In Study 3, only benevolently
sexist women accepted an explicitly group-based protective justification (“It is not safe for
any woman”) for a partner’s imagined opposition to an internship that involved
interviewing criminals. By fusing benevolence with dominance, protective paternalism can
lead women (especially those who are high on benevolent sexism) to accept restrictions.
It’s For Your Own Good - 3
It’s for Your Own Good:
Benevolent Sexism and Women’s Reactions to Protectively Justified Restrictions
Restrictions on women’s behavior may be legitimized through protective (not just
hostile) justifications. Women were once excluded from dangerous or demanding
occupations because of their perceived delicacy, a practice that became institutionalized
(Padavic & Reskin, 2002). Although nondiscrimination laws have made these practices
less prevalent, paternalistic protectiveness may be common within familial and romantic
relationships. For example, a husband may take charge of the couple’s finances because it
would be “too stressful” for his wife. Protectively benevolent justifications may be
effective at eliciting compliance compared to no justification (e.g., “Because I say so”) or
hostile justifications (e.g., “Because women are less capable”). However, the backdrop of
paternalism in gender relations may create attributional ambiguity (Crocker, Voelkl, Testa,
& Major, 1991) for female targets of protective restrictions. Indeed, whether a specific act
is “objectively” discriminatory paternalism may be impossible to determine (see Jackman,
1994 or VanDeVeer, 1986, for detailed discussions). Rather, our concern is how women
resolve this ambiguity. When a man tries to persuade a woman that an activity or job
would be too stressful or dangerous for her, does she perceive his motives as benevolent or
controlling, the restriction as reasonable or discriminatory? The current studies focus on
the personal and situational factors related to how women react to protective restrictions.
That protective justifications are particularly likely to target women is consistent
with recent theory and research on paternalistic sexism (Glick & Fiske, 1996; Jackman,
1994). Ideologically, paternalism is manifested by subjectively affectionate attitudes that
are nevertheless condescending and reinforce women’s lower status. Behaviorally,
It’s For Your Own Good - 4
paternalism includes “positive” behavior (e.g., praise) that helps to keep women in lower
status positions (Vescio, Gervais, Snyder & Hoover, 2005). Paternalist discrimination
creates ambiguity partly because it does not fit people’s naïve prototype of discrimination
as an overtly hostile act (e.g., see Barreto & Ellemers, 2005).
Cleary, women’s ideological acceptance or rejection of paternalism may be an
important factor in their reactions to ostensibly protective restrictions. Ambivalent sexism
theory (Glick & Fiske, 1996, 2001) posits that gender-traditional ideology is manifested in
both hostile sexism (HS, which justifies men’s power by demeaning women’s competence)
and benevolent sexism (BS). BS is fostered by men’s intimate interdependence on women
and includes protective paternalism (e.g., “Women ought to be protected and provided for
by men”), idealization of women (e.g., as “pure” and delicate), and celebration of
heterosexual intimacy (e.g., “Every man needs a woman he adores”). Cross-national
comparisons show that BS is positively correlated with HS and predicts structural
indicators of gender inequality, yet BS also predicts subjectively favorable stereotypes of
women (Glick et al., 2000, 2004). Thus BS fulfills the hallmark of paternalism, a
“positive” attitude toward a subordinate group that nevertheless reinforces their lower
status and power (Jackman, 1994).
Subordinates can find paternalism hard to resist (Jackman, 1994). Indeed,
compared to men, women tend to be equally or even more accepting of BS whereas they
generally reject HS (Glick et al., 2000, 2004; see also Kilianski & Rudman, 1998). Further,
women’s acceptance of BS may be a lynchpin for their acquiescence to other system-
justifying beliefs. Glick et al. (2004) found that correlations between BS and other gender-
traditional beliefs were stronger among women than among men. In a priming study, Jost
It’s For Your Own Good - 5
and Kay (2005) found that women (but not men) who were exposed to BS items
subsequently rated society as fairer. These findings suggest that women who accept BS are
willing to trade some degree of independence for BS’s promise of men’s protection. Thus,
it seems reasonable to posit that women who endorse BS will be more receptive to
protectively justified restrictions.
Even for women who are high in BS, however, acceptance of a protectively
justified restriction may depend on who attempts to impose it. Theoretically, BS is rooted
in intimate interdependence between the sexes. Many BS scale items either explicitly or
implicitly point to male intimates as the source of paternalistic protection (e.g., “A good
woman ought to be set on a pedestal by her man” or “Men should be willing to sacrifice
their own well being in order to provide financially for the women in their lives”).
Therefore, women who are high in BS may be especially receptive to protectively justified
restrictions from intimate male partners. Part of being “set on a pedestal” is accepting a
more passive role, while allowing one’s male partner to “take care of things.” For high BS
women, this may include small symbolic acts of chivalry (e.g., men pulling out chairs) that
low BS women reject. By contrast, protective justifications offered by a non-intimate (with
whom women are not intimately interdependent) may ring hollow. As Jackman (1994)
notes, paternalism derives its persuasive power by infusing acts of dominance with
affection, making it even more effective when initiated by an intimate.
Although low (as opposed to high) BS women may be more likely to view
protective acts as potentially discriminatory, even low BS women may accept protective
restrictions if a male actor’s concerns seem particularly credible, such as when the situation
is truly dangerous and the male actor is a trusted other. For example, given that women are
It’s For Your Own Good - 6
statistically more likely to be victims of sexual violence, low (as well as high) BS women
may view a boyfriend’s insistence that he walk her home after dark favorably rather than
as a restriction of independence. Thus, the general expectation that intimates ought to (and
are) concerned with one’s well-being (Clark & Mills, 1979, 1993) may lead low BS
women to be receptive to protective restrictions imposed by an intimate, but only in
situations that are plausibly dangerous. In sum, we expect that high BS women will
generally be more accepting and low BS women more skeptical of protective restrictions,
but that situational factors will matter. When the circumstance is dangerous, a male
intimate’s protective concern is likely to be viewed as justified.
The manner in which a protective justification is couched may also matter. Given
that high BS women endorse such statements such as “Women ought to be cherished and
protected by men,” they may be receptive to protective justifications that explicitly appeal
to their status as women. For example, if a man justifies a restriction by stating that an
activity “is not safe for any woman,” women who are high (as opposed to low) in BS may
be more likely to acquiesce. Because it invokes their group status and implies that all
women should be restricted in this manner, such an explicitly group-based justification
may be a red flag to low BS women that tags the restriction as discriminatory. A more
personalized protective justification (e.g., “I worry that this isn’t safe for you”), however,
may help to override such suspicions. Emphasizing that the restriction stems from an
intimate’s personal concern may circumvent the perception that it has anything to do with
group status (see Jackman, 1994). Thus, a personalized justification may be able to
puncture the skepticism of low BS women, leading them to react as positively as high BS
women. By contrast, both low and high BS women can be expected to reject explicitly
It’s For Your Own Good - 7
hostile justifications for restrictions. Even though high BS women are more accepting of
traditional roles, they react negatively to women being demeaned (e.g., through hostile
sexist humor; Greenwood & Isbell, 2002).
Overview of Studies and Predictions
Three studies examined women’s reactions to protectively-justified restrictions.
Study 1 used a hypothetical scenario in which a man restricts a woman from driving on a
long trip. This tested predictions that women, regardless of their BS scores, would reject
(a) a hostile justification (“Women are poor drivers”) and (b) a non-intimate male
acquaintance’s attempt to invoke a protective justification (“It will make you tired and
stressed”). However, we expected that (c) women who are high (as compared to low) in BS
would be particularly receptive to a protectively-justified restriction by a male intimate.
In Study 2, we constructed a realistic and plausibly dangerous situation where
female social science students believed they could participate in an internship counseling
wife-abusers and rapists. Each participant received a handwritten (but scripted) note from
her boyfriend opposing her participation. This study tested predictions that (a) when
attributional ambiguity is high (the boyfriend offers no justification), women who are high
(but not low) in BS are more likely to assume benevolent motives, but that (b) given the
potential danger in the situation, even women low in BS would accept a personalized
benevolent justification (“I am worried that it won’t be safe for you”).
Study 3 used a hypothetical scenario in which male and female law students
imagined being offered a legal internship that involved interviewing violent criminals, with
their intimate partner opposing their participation. This study expanded on Study 2 by
examining men’s reactions and adding (for women only) a condition using a group-based
It’s For Your Own Good - 8
protective justification (the situation “is not safe for any woman”). This tested predictions
that (a) men, as members of the dominant gender, might accept a female partner’s
restriction as positively motivated because they do not have the alternative explanation that
the opposition is “gender discrimination” toward them and (b) only women who are high
(not low) in BS would be receptive to a group-based protective justification.
In all three studies, the dependent variables included positive responses (e.g.,
feeling protected), negative responses (e.g., feeling dominated), and whether the act was
discriminatory. Although we expected that positive and negative responses might form a
bi-polar scale, factor analyses showed that participants distinguished between them. Thus,
we report separate analyses for positive and negative reactions in each study.
STUDY 1
In Study 1 women reacted to a hypothetical scenario in which a man attempted to
take over all of the driving on a trip. Type of justification and relationship with the male
actor were varied in a 2 (Justification: hostile, protective) x 2 (Relationship: co-worker,
husband) design.
Participants
A community sample of 141 women from southern Spain were recruited by having
each student in a social psychology class at the Universidad de Granada ask one woman
(typically a mother, grandmother, or aunt) to fill out a survey “examining how women
perceive interactions between women and men.” Students received extra credit for the
completed surveys. The average age of participants was 40; 43% were not in the paid work
force; 80% were living with a male partner; 72% had children; 25% had completed only
It’s For Your Own Good - 9
primary school, 40% had completed only high school or vocational school, and 35% had a
university degree.
Procedure
Participants read a scenario in which a man insisted that he do all of the driving on
a long trip. Depending on random assignment: (a) the man offered either a hostile
justification (women do not drive as well as men) or a protective justification (driving a
long way would be tiring and stressful for the woman) and (b) the man was described as a
co-worker (traveling with her on a business trip) or the woman’s husband (traveling with
her on a vacation).
Dependent measures. Participants completed dependent measures that had been
developed from a pilot study on open-ended responses to discriminatory scenarios.
Multiple items assessed positive and negative feelings about and construals of the man’s
behavior using a 1 (not at all) to 7 (extremely) scale. In addition, a single item asked
whether participants viewed the behavior as discrimination against her “as a woman.”
Because we expected positive and negative responses to yield negatively correlated factors,
we performed a maximum likelihood factor analysis with an oblimin rotation. This
analysis yielded two negatively correlated factors (r = -.38) with eigenvalues greater than
one. Factor 1 (eigenvalue = 5.60, accounting for 43% of the variance) included positive
feelings (protected, cherished, important to him, flattered) and positive attributions (he
feels responsible for me, he is worried about me), which were averaged to form a positive
responses scale (α = .88). Factor 2 (eigenvalue = 1.73, accounting for 13% of the variance)
included both negative feelings (despised, angry, inferior, furious, dominated) and negative
attributions (he has no confidence in me, he is selfish), which were averaged to form a
It’s For Your Own Good - 10
negative responses scale (α = .80). The positive and negative response scales were
moderately correlated with each other (r = -.50, p < .01) and with perceived discrimination
(rs = -.48 and .61, respectively, both ps < .01).
ASI. After completing the dependent measures, participants answered the 22-item
Spanish-language version of the Ambivalent Sexism Inventory (ASI; Glick & Fiske,
1996).
1
Earlier studies in Spain have established its reliability and validity (Expósito,
Moya, & Glick, 1998; Glick et al., 2000, 2004). The ASI comprises two 11-item subscales
that measure HS and BS. All items are statements to which participants respond on a 0
(strongly disagree) to 5 (strongly agree) scale. Sample HS items include: “Women are too
easily offended,” “Feminists are seeking for women to have more power than men,” and
“Most women fail to appreciate fully all that men do for them.”
2
Sample BS items include:
“No matter how accomplished he is, a man is not truly complete as a person unless he has
the love of a woman,” “In a disaster, women ought to be rescued before men,” and “Many
women have a quality of purity that few men possess.” Consistent with prior research, the
HS (α = .85) and BS (α = .80) scales were positively correlated (r = .53, p < .01).
Results
Because the independent variables included categorical (relationship and
justification conditions) as well as continuous variables (HS and BS scores), we conducted
multiple regression analyses using dummy-coded variables for type of justification
(protective = 0, hostile = 1) and type of relationship (partner = 0, co-worker = 1).
Following procedures recommended by Aiken and West (1991), we entered the dummy
variables in Step 1 along with centered BS and HS scores. Step 2 included all 2-way
interactions, followed by 3-way interactions in Step 3. Because there was insufficient
It’s For Your Own Good - 11
power to examine 4-way interactions, we did not proceed further. For reports of regression
contrasts involving ASI scores, we used comparisons at +1 and -1 standard deviations.
Positive Responses
For positive reactions (e.g., feeling protected), there were several significant main
effects. Women reported more positive reactions to being restricted from driving if they
were: (a) high (rather than low) in BS, t(122) = 2.90, β = .29, p < .01; (b) provided with a
protective as opposed to a hostile justification, t(122)= 2.10, β = -.17, p < .01; and (c)
restricted by a husband as compared to a co-worker, t(122) = 3.56, β = -.29, p < .001.
However, these effects and related 2-way interactions must be interpreted in terms of
significant 3-way BS x relationship x justification interaction, t(112) = 1.90, β = .28, p <
.05, illustrated in Figure 1.
The 3-way interaction revealed that women who scored high (rather than low) on
BS reported especially positive feelings when an intimate partner used a protective
justification to forbid her from driving. Tests of simple slopes (Aiken & West, 1991)
revealed that when a co-worker discriminated, the type of justification offered did not
affect women’s responses; this held true both for women low in BS, t(57) = .72, β = .13,
ns, and high in BS, t(57) = .98, β = -.20, ns. In contrast, when a husband discriminated,
women high in BS reported more positive feelings in response to a protective as compared
to a hostile justification, t(60) = 3.71, β = .54, p < .001, but women low in BS did not show
this effect, t(60) = 1.14, β = .16, ns.
In sum, it was only when a husband provided a protective (rather than hostile)
justification for trying to restrict the woman’s behavior that women who were high (but not
low) in BS reported more positive feelings (relative to all other conditions). In comparison,
It’s For Your Own Good - 12
when a co-worker discriminated, a protective as compared to hostile justification did not
boost positive responses, either for high or low BS women.
Negative Responses
For negative reactions (e.g., feeling angry), there was a significant main effect of
type of justification – women reported less negative reactions in response to a protective as
opposed to a hostile justification, t(122) = 3.58, β = .30, p < .01. However, this effect was
qualified by a justification x relationship interaction, t(116) = 2.81, β = -.34, p < .01. Tests
of the simple slopes showed that when a husband was the source of the restriction, women
reported less negative reactions to the protective as opposed to the hostile justification,
t(67) = 4.15, β = .45, p < .001. In contrast, when a coworker restricted the woman from
driving, the type of justification did not have an effect, t(68) = .83, β =.10, ns. Unlike the
findings for positive reactions, the relationship x justification interaction was not
moderated by BS. No other effects were significant. In sum, a protective as compared to a
hostile justification led to less negative reactions, but only when the restriction was
imposed by a husband, not a co-worker.
Perceived Discrimination
Was the man’s action perceived as discriminatory? There was a main effect for
justification such that the act was viewed as less discriminatory if a protective as compared
to hostile justification was given, t(122) = 4.18, β = .34, p < .001. More importantly, there
was a 3-way justification x relationship x BS interaction, t(112) = 2.06, β = .32, p < .05.
This interaction revealed that only women who were high in BS viewed the act as non-
discriminatory and only if it was committed by a partner who offered a protective
justification.
It’s For Your Own Good - 13
More specifically, tests of simple slopes revealed that when the restriction was
imposed by a co-worker, the type of justification did not significantly affect perceived
discrimination either for low BS women, t(58) = 1.64, β = -.30, ns, or for high BS women,
t(58) =.98, β = -.20, ns. When the restriction was imposed by a husband, low BS women’s
perception of discrimination was not significantly affected by the justification, t(64) =
1.82, β = -.29, ns; however, women who were high in BS perceived the act as non-
discriminatory in the protective as compared to hostile justification condition, t(64) = 3.88,
β = -.62, p = .001.
In sum, the type of justification diminished women’s perceptions of the man’s
restriction as discriminatory only for benevolently sexist women and only when the
restriction was imposed by a husband who gave a protective justification. Thus, the
perceived discrimination measure mirrored (in inverse fashion) women’s ratings of
positive responses to the imagined restriction.
Discussion
Study 1 demonstrated that only benevolently sexist women reported positive
reactions to a scenario in which a husband (but not a co-worker) gave a protective (versus
hostile) justification for refusing to let them take the wheel on a long car trip. For these
women, a husband’s protectively justified restriction was not viewed as discriminatory. By
contrast, high BS women (like their low BS counterparts) did not react positively when a
co-worker was the source of the protective restriction, viewing it instead as discriminatory.
While a consistent picture emerged for measures of positive reactions and judgments of
discrimination, the negative reaction measure showed simpler effects. Women (regardless
of BS scores) reacted less negatively to a protective (versus hostile) justification, but only
It’s For Your Own Good - 14
when the source was a husband not a co-worker. It may be that the absence of more
complex effects on negative reactions is due to people’s general reluctance to impute
negative motives to other individuals (Sears, 1983).
In sum, women (regardless of their BS scores) did not view a protective
justification from a non-intimate as credible. Low BS women remained skeptical even
when the source of the protectively justified restriction was a husband, but high BS women
were (as expected) particularly prone to react positively when an intimate male partner
offered a protective justification for restricting their behavior.
STUDY 2
Several shortcomings in Study 1 were compensated for in Study 2. First, Study 1’s
scenario methodology presented women with a hypothetical rather than a realistic
situation. Second, Study 1’s hypothetical situation was not plausibly dangerous; the man
was “protecting” the woman merely from becoming tired or stressed on a long drive.
Third, the behavior women imagined being restricted from was also relatively trivial and
perhaps not one they were particularly interested in performing. Finally, Study 1 failed to
include a no-justification condition. Such a condition is of special interest because it
maximizes ambiguity about the male actor’s motives and might therefore increase the
likelihood that individual differences in endorsement of BS would predict women’s
reactions (see Snyder & Ickes, 1985).
Thus, Study 2’s methodology was designed to be highly realistic, presenting
women with a desirable opportunity that also entailed a plausible danger. Specifically,
female social science students were given an apparent practicum opportunity counseling
rapists and wife abusers. Also, rather than a hypothetical restriction, women responded to
It’s For Your Own Good - 15
apparently real opposition from their actual boyfriend (who had accompanied them).
Finally, we added a “no justification” control condition that maximized attributional
ambiguity about the boyfriend’s motives. We expected that women high (versus low) in
BS would be more likely to infer protective motives when their boyfriends opposed the
practicum but failed to explain why. However, given the danger inherent in the practicum,
we reasoned that even women who are low in BS would react positively to a personalized
protective justification (“I am worried that you won’t be safe”).
Methods
Participants
A total of 116 female students (averaging 19.5 years old) in Granada, Spain
participated. All were majoring in psychology, sociology, or education and had indicated
on a pre-selection form that they currently had a steady romantic partner. On average,
women had been in the relationship for 1.64 years.
3
Procedure
Each participant was contacted to arrange a time when she and her partner could
attend a session for class credit (at this point participants had not been informed of any
practicum opportunity). Each session was run with up to three couples. When each pair
arrived, the partners were immediately taken to separate rooms.
The female partner first completed the ASI. Meanwhile, the male partner was told
the true nature of the study and asked if he would copy, in his own hand-writing, the
feedback that had been randomly assigned to his partner. All male partners consented to do
so. While the male partner copied the assigned message, the experimenter informed the
female partner that she might be eligible for a counseling practicum working with men
It’s For Your Own Good - 16
who had been convicted of rape or wife abuse. An information packet explained that she
had been asked to bring her partner because “it is important for those who participate to be
supported by their partners and to ensure that participation will not negatively affect your
relationship.” Women then rated their interest in the practicum should they be selected.
The level of interest was generally very high (M = 5.63 on a 1 to 7 scale) and did differ
between conditions, F(2, 100) = 1.84, ns.
Each woman was told that her male partner had also been informed about the
opportunity and had written his reaction to it. The women were then given the note their
partner had copied. In all cases the note indicated the partner’s strong opposition. In the no
justification condition it stated “I would tell her that she cannot do this. I think it is a bad
idea. Even if she thought she really wanted to do it, I would get her not to. I am sure I
could convince her.” In the hostile justification condition, the note read “I would tell her
that she cannot do this. I think it is a bad idea because this is not a situation in which a
woman, compared to a man, could be effective. Even if she thought she really wanted to do
it, I would get her not to because a woman shouldn’t get in these situations when there are
men that can handle it much better. I am sure I could convince her.” In the personalized
protective justification condition, the note stated: “I would tell her that she cannot do this. I
think it is a bad idea because it would not be safe for her and I think she would find it to be
really stressful. Even if she thought she really wanted to do it, I would get her not to
because I would be very worried about her safety and well-being in this situation. I am sure
I could convince her.”
The dependent measures were the same as those in Study 1: positive responses (α =
.84), negative responses (α = .81), and whether the partner’s opposition was discrimination
It’s For Your Own Good - 17
against her as a woman. The positive and negative response scales were modestly
correlated with each other (r = -.27, p < .01) and each was correlated with perceived
discrimination (rs = -.39 and .53, respectively, both ps < .01).
Ethical considerations and debriefing. Considerable care was taken to prevent
psychological harm. Both partners were initially told they could choose not to answer any
questions and end their participation at any time without penalty. Male partners were
informed of the reasons for the deceptive feedback and reminded they could refuse to
participate with no penalty for their girlfriend. Female partners were immediately and
extensively debriefed to explain that the partner’s feedback was false and randomly
assigned.
Results
Control Variables
We examined whether the length of the women’s relationships or their interest in
the internship correlated with ASI scores or reactions to the partner’s feedback. Length of
relationship was not related to BS or HS scores (both rs < .14, ns), nor did it predict
positive or negative reactions (including perceived discrimination) to the boyfriend’s
opposition (all rs < .14, ns). Degree of interest in the internship was also unrelated to
women’s HS and BS scores (both rs < .11, ns), but did predict less positive reactions (r = -
.41, p < .01), more negative reactions (r =.21, p < .05), and greater perceived
discrimination (r = .26, p < .01) in response to the boyfriend’s feedback. However,
regression analyses in which this variable was included showed that it did not moderate the
effects reported below.
Analysis Strategy
It’s For Your Own Good - 18
We conducted multiple regression analyses to assess women’s reactions as a
function of type of justification (hostile, no justification, or protective) and women’s BS
and HS scores. We created dummy variables to yield two contrasts: the first contrast
(hostile = 1, no justification = 0, protective = 0) compared the hostile and no justification
conditions; the second contrast (hostile = 0, no justification = 0, protective = 1) compared
the protective and no justification conditions. Together, these contrasts (entered in Step 1)
tested the justification main effects. In Step 2, we entered centered BS and HS scores,
followed by the interactions between the contrasts and the ASI scales in Step 3. We did not
proceed beyond these 2-way interactions.
Positive Responses
There was a significant main effect for condition, F(2, 113) = 7.52, p < .001.
Women responded more positively to no justification as compared to a hostile justification,
t(113) = 2.95, β = -.30, p < .01. However, the protective justification did not elicit more
positive responses than no justification, t(113) = 1.25, β = .12, ns. More importantly, BS
generally predicted more positive responses, t(106) = 3.05, β = .55, p < .01, and interacted
with the type of justification, t(106) = 2.22, β = -.32, p < .05 for the hostile justification;
t(106) = 2.44, β = -.35, p < .05 for the protective justification. These interactions are
graphed in Figure 2.
Tests of simple slopes showed that women high in BS reported significantly more
positive feelings in the no justification than in the hostile justification condition, t(106) =
3.48, β = .59, p < .001, but they reacted equally positively to no justification as they did to
a protective justification, t(106) = 1.24, β = -.19, ns. In contrast, women low in BS
exhibited an equal (and low) degree of positive reaction in the no justification and hostile
It’s For Your Own Good - 19
justification conditions, t(106) = .17, β = -.03, ns, but evinced significantly more positive
reactions in the protective as compared to the no justification condition, t(106) = 2.20, β =
.37, p < .05. Contrasts within conditions revealed that BS was significantly related to
positive reactions only within the no justification condition, t(36) = 3.42, β = .58, p < .001,
whereas in the explicit justification conditions (whether hostile or protective) BS scores
did not predict women’s reactions, both ts (36) < .25, βs < .03, ns.
In sum, women who were high in BS reported feeling just as positively when their
partner offered no justification for restricting them as when he provided a personalized
protective justification. In contrast, women low in BS treated no justification similarly to a
hostile justification. These differences between high and low BS-scoring women
disappeared, however, when the partner provided a personalized protective rationale – in
this condition, low BS women responded just as positively as high BS women.
Negative Responses
The degree of women’s negative responses to their boyfriend’s opposition was
affected by the type of justification he offered. The first step in the regression revealed a
main effect for condition, F(2, 113) = 10,45, p < .001. Main effects were significant for
both the hostile justification, t(113) = 2.78, β = .27, p < .01, and the protective justification,
t(113) = -2.62, β = -.25, p <.05, were significant. These effects indicated that women
reacted more negatively when their boyfriends used a hostile (versus no) justification and
less negatively when they used a protective (versus no) justification. There were no other
significant effects – neither BS nor HS moderated the main effects, all ts < 1.90, βs < .25,
ns. Thus, compared to the no justification condition, when the boyfriend gave an explicitly
It’s For Your Own Good - 20
hostile justification, women responded more negatively and when he gave a protective
justification women responded less negatively.
Perceived Discrimination
Did women perceive their partners’ opposition as gender discrimination? This
measure was strongly affected by condition, as indicated by a significant main effect, F(2,
113) = 23.46, p < .001. Not surprisingly, the hostile justification elicited a significantly
greater likelihood of labeling the partner’s reaction as discriminatory (M = 5.47) as
compared to the no justification condition (M = 3.38), t(113) = 4.81, β = .44, p < .001.
There was a nonsignificant trend for women to be less likely to label their partner’s
reaction as discriminatory when he gave a protective justification (M = 2.59) as compared
to no justification, t(113) = -1.84, β = -.17, p <.07. No other effects (including
interactions) were significant. Thus, only an explicitly hostile justification increased the
likelihood that the boyfriend’s restriction was perceived as discriminatory as compared to
when no justification was provided.
Discussion
When women were faced with an apparent attempt by their own romantic partner to
restrict them from a practicum opportunity, a personalized protective (as compared to no)
justification elicited more positive and less negative responses (and was marginally less
likely to be viewed as discriminatory). Importantly, the personalized protective
justification eliminated individual differences in positive reactions based on women’s BS
scores (i.e., low BS women reacted just as positively to the personalized protective
justification as high BS women). However, when the partner offered no justification for his
opposition, individual differences in women’s BS mattered – high BS women responded to
It’s For Your Own Good - 21
no justification just as positively as they did to a personalized protective justification. In
contrast, low BS women responded to the no justification condition with less positive
responses that were similar to how they reacted to a hostile justification. Thus, the
ambiguous no justification condition appeared to provide fertile ground for women to
interpret the partner’s motives, allowing their own ideologies to color their construals.
STUDY 3
Although Study 2 answered a number of questions raised by Study 1, it left two
important issues unexamined: (a) would men react more or less positively than women to a
partner’s protectively justified restriction and (b) would women who are high (as compared
to low) in BS react as positively to a protective justification that is explicitly group-based
(e.g., “It is not safe for any woman”)? In terms of men’s reactions, we have argued that
because they belong to the dominant gender and are not typically subjected to paternalism,
they are unlikely to think that a protectively justified restriction reflects “gender
discrimination.” In other words, while women have two potential explanations for a
partner’s protective restriction (personalized concern versus group-based discrimination),
the notion that a partner’s opposition is discriminatory is unlikely to occur to men.
From this standpoint, men (as compared to women) might have more positive
reactions to a partner’s opposition because the only likely explanation is “she is concerned
about me.” Although they could interpret her opposition as an effort to exert control
(prompting some negativity), gender stereotypes and roles mitigate against “selfish”
explanations of a woman’s opposition to a male partner taking a dangerous internship.
First, women (unlike men) are conventionally viewed as benefiting from (and being in no
way diminished by) their intimate male partner’s career success, even if it is greater than
It’s For Your Own Good - 22
their own. Second, gender stereotypes cast women as nurturing and emotionally
vulnerable, making it more plausible that their concern comes from caring for the partner.
In short, because they are unburdened by a history of paternalistic restriction, men (as
compared to women) may be: (a) more likely assume positive motives for a female
partner’s attempt to restrict them from a potentially dangerous internship and (b) less likely
to view such a restriction as gender discrimination.
The second issue Study 3 addressed is women’s reactions to a group-based
protective justification. In Studies 1 and 2, the protective justification (unlike the hostile
one) did not invoke generalizations about women as a group. Study 2 showed that a
personalized protective justification (e.g., “It is not safe for you”) can override skepticism
about a male partner’s motives, even for low BS women. But how would women who are
low versus high in BS react to a group-based justification (e.g., “It is not safe for a
woman”)? Because women who are low in BS reject paternalistic sexism, we expected that
they would react less positively to a group-based justification. In contrast, because such a
justification is consistent with benevolently sexist ideology, we expected that high BS
women would respond relatively positively to a group-based protective justification,
perhaps even treating it similarly to a personalized protective justification.
Study 3’s dual purpose dictated an unbalanced experimental design. Specifically,
while we added male participants to compare men’s and women’s reactions to past
conditions, the new “group-based” condition did not seem appropriate for men. Men are
not normally subjected to paternalism; thus, a group-based protective justification (e.g., “It
is not safe for a man”) is likely to be seen as a non sequitur by male participants. To
compare across male and female participants part of the design was fully crossed,
It’s For Your Own Good - 23
comprising a 2 (Sex of Participant) x 2 (Justification: no justification, personalized
protective justification) design. To test how women would react to an explicitly group-
based protective justification, we added this condition for women only.
Methods
Participants
The participants included 123 male and 168 female law students from the
Universidad de Granada Law School. The average age of participants was 20.6 years and
about 70% reported currently being in a romantic relationship; the median duration of
relationships among these respondents was 21 months.
Procedure
Participants were given a scenario during a law school class that asked them to
imagine being offered a legal studies internship that reexamines evidence in cases where a
convict claims to be innocent. Interns were said to “conduct in-depth interviews with
people who have been convicted of violent crimes, including murder.” Participants were
asked to imagine that they were excited about the opportunity. However, “when you
explain the summer internship and all that it involves, your romantic partner strongly
opposes your participation saying ‘I really think this is a bad idea. You shouldn’t do this!’”
Depending on random assignment, some participants received only the statement of
opposition (listed above) without further explanation (the no justification condition).
Others received the opposition statement plus a personalized protective justification: “I
don’t think it’s good for you to be exposed to people who have committed violent crimes.
It seems like a stressful and dangerous situation that wouldn’t be safe for you.” For female
participants only, some received the opposition statement plus a group-based protective
It’s For Your Own Good - 24
justification: “I don’t think it’s good for any woman to be exposed to people who have
committed violent crimes. It seems like a stressful and dangerous situation that wouldn’t
be safe for a woman” (italics added here to highlight the differences from the personalized
protective justification).
Participants completed the same measures used in the prior studies: positive
responses (α = .85), negative responses (α = .82), whether the partner’s opposition
constituted gender discrimination, and the ASI. The positive and negative response scales
were moderately correlated with each other (r = -.47, p < .01) and with perceived
discrimination (rs = -.36 and .56, respectively, both ps < .01).
Results
As with the prior studies, we used multiple regression to analyze the data. Because
the design was not fully crossed, however, each dependent variable was analyzed in two
ways. First, to test comparisons between male and female participants we dummy-coded
sex of participant (0 = female, 1 = male) and type of justification (0 = no justification, 1 =
personalized protective). These variables were entered in a regression along with centered
HS and BS scores (with main effects in Step 1, 2-way interactions in Step 2, and 3-way
interactions in Step 3). The first analysis partially replicates Study 2, but with the addition
of male respondents.
Second, to test how women reacted to the new group-based protective justification,
we examined women only, comparing their responses to the new group-based justification
and each of the other justification conditions using two dummy-coded contrasts: (a) 0 =
group-based, 1 = no justification, 0 = personalized protective and (b) 0 = group-based, 0 =
no justification, 1 = personalized protective. These contrasts were entered in Step 1 along
It’s For Your Own Good - 25
with centered HS and BS scores, followed by the 2-way interactions at Step 2. This second
analysis focused on women’s reactions to a type of justification (group-based) we had not
previously examined.
Positive Responses
Comparing women and men. Did men react more or less positively than women to
a protective restriction? Results are illustrated in Figure 3. The regression analyzing
women’s and men’s responses revealed a marginally significant main effect for sex of
participant, such that men reacted more positively than women to the partner’s opposition,
t(229) = 2.09, β = .13, p < .06. Two other main effects were significant: (a) respondents
who were high in BS evinced more positive responses, t(229) = 2.09, β = .14, p < .05, and
(b) participants responded more positively in the personalized protective versus no
justification condition, t(223) = 3.88, β = .24, p = .01. These main effects, however, must
be interpreted in light of a sex of participant x type of justification x BS interaction, t(221)
= 2.03, β = .28, p < .05. Tests of simple slopes revealed that high BS men responded
significantly more positively to a personalized protective justification than to no
justification, t(117) = 2.89, β = .32, p < .01, whereas low BS men responded equally
positively to both conditions, t(117) = -.02, β = .00, ns. Indeed, high BS men in the
personalized condition appear to account for men’s overall tendency to respond more
positively than women (in all other conditions, men’s scores were virtually identical to
women’s). For female participants, results were similar to Study 2. In the no justification
condition, there was a marginally significant tendency for high (versus low) BS women to
react more positively, t(55) = 1.94, β = .25, p < .06 , but this tendency disappeared in the
personalized protective justification condition, t(54) = .76, β = .10, ns. Across conditions,
It’s For Your Own Good - 26
women who were high in BS responded equally positively to no justification and a
personalized protective justification, t(109) = 1.64, β = .24, ns, whereas low BS women
responded less positively to the former than to the latter, t(109) = 2.96, β = .34, p < .01.
Women’s reactions to a group-based justification. The female-only analysis, which
compared women’s responses to the group-based protective justification to the other two
conditions, can also be seen in Figure 3. Women reacted more positively to the
personalized as compared to group-based protective justification, t(163) = 3.70, β = .32. p
< .01. In contrast, there was no difference in positive reactions between the group-based
protective and no justification conditions, t(163) = .21, β = .02, ns. There was also a
significant BS main effect such that high BS women generally evinced more positive
reactions across all conditions, t(163) = 2.61, β = .20, p = .01. Finally, there was a
marginally significant interaction between BS and the group-based versus personalized
protective justification comparison, t(158) = 1.93, β = -.21, p = .055. More specific
contrasts showed that high BS women reacted only marginally significantly more
positively in the personalized protective as compared to group-based protective
justification condition, t(164) = 1.68, β = .20, p < .10. In contrast, women who were low in
BS responded significantly more positively to the personalized protective justification than
they did to the group-based justification, t(164) = 3.98, β = .37, p < .01. Also, within the
group-based justification condition, women’s BS score predicted significantly more
positive reactions to the partner’s opposition, t(53) = 2.93, β = .37, p < .01, whereas (as
reported above) this was not the case when a personalized protective justification was used.
Overall, men reacted slightly more positively than women to the partner’s
restriction, but this was due to high BS men’s especially positive response to a
It’s For Your Own Good - 27
personalized protective justification. For women, reactions to the group-based protective
justification were similar to their responses to no justification – women’s BS scores
predicted whether the group-based justification was perceived positively. In comparison to
a personalized justification, women responded significantly less positively to a group-
based protective justification, but this was particularly true for women who were low in
BS. It was only in the personalized protective condition that low as well as high BS women
reacted equally positively.
In general, women’s positive responses replicated effects from Study 2. First, when
the reason for the partner’s opposition was ambiguous (no justification), high BS women
assumed benevolent motives, whereas low BS women did not. Second, a personalized
protective justification overrode differences in BS, with low BS women reacting just as
positively as high BS women. The new group-based protective justification elicited a
positive reaction from women who were high but not women who were low in BS. For
men, the only significant effect was that high (but not low) BS men showed more positive
reactions to a personalized protective justification (as compared to no justification).
Negative Responses
Comparing women and men. The regression that explored women’s and men’s
negative reactions to the no justification versus personalized protective justification
conditions yielded several main effects. Negative reactions were stronger (a) in the no
justification versus personalized protective justification condition, t(229) = -2.49, β = -.16,
p < .05, (b) for women as compared to men, t(229) = -4.42, β = -.31, p < .01, and (c) for
people who scored higher on HS, t(229) = 3.09, β = .23, p < .05. The latter two effects
were qualified by a highly significant sex of participant x HS interaction that occurred
It’s For Your Own Good - 28
because men (and not women) who scored high in HS showed more negative reactions,
t(223) = 3.71, β = .40, p < .01. No other effects were significant.
Women’s reactions to a group-based justification. For the regression comparing
women’s reactions in the group-based protective justification condition to the other
conditions, the only significant effect was for the contrast between the group-based versus
personalized protective conditions, t(163) = -3.26, β = -.29, p < .01. Women had more
negative reactions to the group-based than the personalized protective justification. In
contrast, the group-based protective justification was not rated differently than the no
justification control, t(163) = -1.20, β = -.11, ns. No other effects were significant.
In sum, reactions were less negative if a partner gave a personalized protective
justification compared to (a) no justification (for both men and women) or (b) a group-
based protective justification (given to women only). Interestingly, men generally showed
less negative responses than women, except for men who were high in HS.
Perceived Discrimination
Comparing women and men. Comparisons of men’s and women’s perceptions of
discrimination are shown in Figure 4. The first analysis (comparing across the sexes for the
no justification and personalized protective justification conditions) revealed a large sex
difference in perceived discrimination: women were much more likely to view their
partner’s opposition as “discrimination against me as a woman” whereas men tended not to
view opposition as “discrimination against me as a man”, t(229) = -5.07, β = -.36, p < .01.
The only other effects were an HS main effect, t(229) = 2.05, β = .15, p < .05, which was
qualified by a sex of participant x HS interaction, t(224) = 3.42, β = .38, p < .01. This was
similar to the interaction that occurred with negative responses – it was driven by men who
It’s For Your Own Good - 29
were high in HS being more likely to rate the partner’s opposition as “discrimination”
(whereas women’s HS scores did not affect perceived discrimination).
Women’s reactions to a group-based justification. For the regression that examined
only women’s perceptions of discrimination, both condition contrasts were significant.
Specifically, women were more likely to rate the group-based as compared to personalized
protective justification as “discriminatory,” t(163) = -2.76, β = -.24, p < .01. Similarly,
women also viewed the group-based protective justification as more discriminatory than
when no justification was given, t(163) = -2.97, β = -.26, p < .01. No other effects were
significant.
Overall, men (compared to women) were unlikely to view a partner’s opposition to
a potentially dangerous internship as gender discrimination (only high HS men showed any
propensity to do so). For women, the group-based protective justification (which
specifically referred to women as a group) was especially likely to be viewed as
discriminatory (in comparison to both the personalized protective and no justification
conditions).
Discussion
Study 3 addressed two new issues. First, would men react more or less positively
than women to a protective restriction? Men reacted marginally significantly more
positively and significantly less negatively to a partner’s restriction than women did. The
strongest effect, however, was that men (averaging about a 2 on a 7-point scale) were
much less likely than women (averaging about a 4) to view the partner’s opposition as
discriminatory. In other words, men did not appear to use “gender discrimination” (as
It’s For Your Own Good - 30
opposed to “genuine worry about my safety”) as an alternative attribution about their
partner’s imagined opposition.
The second issue was how women would react to a specifically group-based
protective justification. We found that high BS women reacted more positively to the
justification that the situation was unsafe for “any woman” than low BS women. This
occurred even though both high and low BS women saw the group-based justification as
more discriminatory than any other justification. As in Study 2, only the personalized
protective justification had the power to erase differences between the reactions of low and
high BS women.
There were two unexpected findings concerning men. First, high (as compared to
low) BS men evinced especially positive responses to a personalized protective
justification. This may have occurred because BS is strongly correlated with benevolence
toward men (Glick & Fiske, 1999; Glick et al., 2004), which advocates that women “take
care of their men” through maternalistic fussing (e.g., catering to men’s needs at home).
The female partner’s protective opposition may have been interpreted by these men as
consistent with such maternalism.
Second, men who were high (as compared to low) in HS evinced more negative
reactions to a partner’s opposition (regardless of how she justified it) and were more likely
to rate such opposition as “discrimination against me as a man.” This is consistent with
HS’s adversarial view of women as controlling (e.g., one HS item asserts “Once a woman
gets a man to commit to her, she usually tries to put him on a tight leash”). It is also
consistent with findings that members of dominant groups who endorse system-justifying
ideologies are more likely to perceive negative outcomes from members of subordinate
It’s For Your Own Good - 31
groups as “reverse discrimination” (Major, Gramzow, McCoy, Levin, Schmader, &
Sidanius, 2002). Thus it is not surprising that high HS men were more resentful. It is
important to note, however, that this effect occurred in the context of a general reluctance
on the part of men to label a partner’s opposition as discrimination.
General Discussion
Together, these studies suggest that protective justifications have the power to elicit
positive reactions to restrictions imposed on women by their romantic partners. Further,
women who endorse BS are particularly prone to assuming a protective justification even
if the partner offers none. They are also receptive to an explicitly group-based justification
that appeals to their vulnerability “as a woman.” Women who reject BS appear to be more
skeptical of men’s motives, but still responded positively to a protective justification when
it was personalized (“I am worried about your safety”), delivered by a romantic partner,
and in response to a plausibly dangerous situation (an internship requiring contact with
violent men). In this situation, the distinction between paternalistic discrimination and
legitimate concern becomes particularly difficult to make. Given the credibility of a
personalized protective justification, low BS women reacted positively, not negatively to a
romantic partner’s imagined restriction. In contrast, when there was no plausible “danger”
(in Study 1’s driving situation), low BS women did not accept a protective justification
from a romantic partner.
Overall, women who endorse BS seem willing to trade some degree of
independence for male protectiveness, even when this protection is explicitly sexist (i.e.,
“because you are a woman”). Interestingly, women’s BS scores did not relate to their
propensity to label a group-based protective justification as discriminatory (both low and
It’s For Your Own Good - 32
high BS women rated this condition as more discriminatory than no justification or a
personalized protective justification). But even though high BS women labeled the group-
based justification as “discrimination” they nevertheless reacted only marginally less
positively to it than to a personalized protective justification. In short, high BS women do
not mind being discriminated against “as a woman” when it is part of a male partner’s
chivalrous protectiveness. Indeed, high BS women not only seem to welcome such
protectiveness, but to expect it from their partners. When attributional ambiguity was
increased by having the partner offer opposition without a justification, benevolently sexist
women assumed that their partner had protective motives.
Although women who reject BS were generally more skeptical of men’s motives, a
personalized protective justification from a romantic partner was elicited positive reactions
from them (at least in plausibly dangerous situations). Of particular interest is how low BS
women reacted to Study 3’s group-based protective justification, which was identical in
every way to the personalized protective justification except for the substitution of “any
woman” for “you.” For low BS women, the group-based justification elicited a reaction
similar to no justification (and, in Study 2, low BS women reacted to the no justification
condition the same as an explicitly hostile justification). By contrast, the personalized
justification led low BS women to react as positively as high BS women.
That the small alteration from “you” to “any woman” had such a dramatic effect
implies, as Jackman (1994) contends, that personalized protective justifications can
effectively overcome resistance to restrictions. Even women who are relatively skeptical of
men’s attempts to curtail their independence appear (in some circumstances) to be
It’s For Your Own Good - 33
susceptible to a romantic partner’s restriction if it is expressed in terms of a “concern for
your safety and well-being.”
The current studies are an initial foray into how women disentangle benevolence
and paternalism in their relationships with men, with limited results. For instance, our
suggestion that the perceived dangerousness of the situation affected women’s reactions is
based on comparisons between studies; a firmer conclusion requires manipulating this
factor within an experiment. Further, although restrictions imposed by a co-worker were
rejected in Study 1 (a driving situation that was not dangerous) it is premature to conclude
that protective justifications are only accepted from male intimates (the focus of Studies 2
and 3). Future studies might examine a wider range of situations and categories of
relationships, as well as other individual difference (e.g., attachment style) and relationship
(e.g., satisfaction) variables that may predict women’s responses.
Conclusion
Protective paternalism is not confined to sexism (see Cuddy, Fiske, & Glick, in
press; Jackman, 1994). However, given the unique structure of gender relations (which
combine male dominance with intimate interdependence between the sexes), protectively
justified restrictions of women are likely to be especially pervasive. By blending affection
with dominance, paternalism makes it particularly difficult for women to distinguish male
intimate’s benign care from manipulative attempts to control their behavior. Women who
endorse paternalism seem to assume that male partners’ motives are benign and welcome
protective restrictions from their partners even when they recognize them as
discriminatory. Women who reject paternalism are considerably more skeptical of men’s
motives, but are at least sometimes willing to accept restrictions as acts of love so long as
It’s For Your Own Good - 34
their partners obscure any influence of gender by justifying opposition as concern for
“your well being” and studiously avoid adding “as a woman.”
It’s For Your Own Good - 35
References
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. Newbury Park: Sage Publications.
Barreto, M., & Ellemers, N. (2005). The burden of benevolent sexism: How it contributes
to the maintenance of gender inequalities. European Journal of Social Psychology,
35, 633-642.
Clark, M. S., & Mills, J. (1979). Interpersonal attraction in exchange and communal
relationships. Journal of Personality and Social Psychology, 37, 12-24.
Clark, M. S., & Mills, J. (1993). The difference between communal and exchange
relationships: What it is and is not. Personality and Social Psychology Bulletin, 19,
684-691.
Crocker, J., Voelkl, K., Testa, M., & Major, B. (1991). Social stigma: The affective
consequences of attributional ambiguity. Journal of Personality and Social
Psychology, 60, 218-228.
Cuddy, A. J. C., Fiske, S. T., & Glick, P. (in press). The BIAS Map: Behaviors from
Intergroup Affects and Stereotypes. Journal of Personality and Social Psychology.
Expósito, F., Moya, M., & Glick P. (1998). Sexismo ambivalente: Medición y correlates
[Ambivalent Sexism: Measurement and correlations]. Revista de Psicología Social,
13(2), 159-169.
Glick, P., & Fiske, S. T. (1996). The Ambivalent Sexism Inventory: Differentiating Hostile
and Benevolent Sexism. Journal of Personality and Social Psychology, 70, 491-
512.
It’s For Your Own Good - 36
Glick, P., & Fiske, S. T. (1999). The Ambivalence toward Men Inventory: Differentiating
hostile and benevolent beliefs about men. Psychology of Women Quarterly, 23,
519-536.
Glick, P., & Fiske, S. T. (2001). Ambivalent sexism. In M. P. Zanna (Ed.), Advances in
Experimental Social Psychology (vol. 33, pp. 115-188). Thousand Oaks, CA:
Academic Press.
Glick, P., Fiske, S. T., Mladinic, A., Saiz, J, Abrams, D., Masser, B., Adetoun, B.,Osagie,
J., Akande, A., Alao, A., Brunner, A., Willemsen, T. M., Chipeta, K., Dardenne,
B., Dijksterhuis, A., Wigboldus, D., Eckes, T., Six-Materna, I., Expósito, F., Moya,
M., Foddy, M., Kim, H-J., Lameiras, M., Sotelo, M. J., Mucchi-Faina, A., Romani,
M., Sakalli, N., Udegbe, B., Yamamoto, M., Ui, M., Ferreira, M. C., & López, W.
L. (2000). Beyond prejudice as simple antipathy: Hostile and benevolent sexism
across cultures. Journal of Personality and Social Psychology, 79, 763-775.
Glick, P., Lameiras, M., Fiske, S. T., Eckes, T., Masser, B., Volpato, C., Manganelli, A.
M., Pek, J., Huang, L., Sakalli-Ugurlu, N., Castro, Y. R., D’Avila Pereira, M. L.,
Willemsen, T. M., Brunner, A., Six-Materna, I, & Wells, R. (2004). Bad but bold:
Ambivalent attitudes toward men predict gender inequality in 16 nations. Journal
of Personality and Social Psychology, 86, 713–728.
Greenwood, D., & Isbell, L. M. (2002). Ambivalent sexism and the dumb blonde: Men's
and women's reactions to sexist jokes. Psychology of Women Quarterly, 26, 341-
350.
Jackman, M. R. (1994). The velvet glove: Paternalism and conflict in gender, class, and
race relations, Berkeley, CA: University of California Press.
It’s For Your Own Good - 37
Jost, J. T., & Kay, A. C. (2005). Exposure to benevolent sexism and complementary
gender stereotypes: Consequences for specific and diffuse forms of system
justification. Journal of Personality and Social Psychology, 88, 498-509.
Kilianski, S. E., & Rudman, L. A. (1998). Wanting it both ways: Do women approve of
benevolent sexism? Sex Roles, 39, 333-352.
Major, B., Gramzow, R. H., McCoy, S., Levin, S., Scmader, T., & Sidanius, S. (2002).
Perceiving personal discrimination: The role of group status and legitimizing
ideology. Journal of Personality and Social Psychology, 82, 269-282.
Padavic, I., & Reskin, B. (2002). Women and men at work (2
nd
ed.). Thousand Oaks, CA:
Sage Publications.
Sears, D. O. (1983). The person-positivity bias. Journal of Personality and Social
Psychology, 44, 233-250.
Snyder, M., & Ickes, W. (1985). Personality and social behavior. In G. Lindsey & E.
Aronson (Eds). Handbook of social psychology (Vol. 2, 3
rd
ed., pp. 883-947). New
York: Random House.
VanDeVeer, D. (1986). Paternalistic intervention: The moral bounds of benevolence.
Princeton, NJ: Princeton University Press.
Vescio, T. K., Gervais, S. J., Snyder, M., & Hoover, A. (2005). Power and the creation of
patronizing environments: The stereotype-based behaviors of the powerful and
their effects on female performance in masculine domains. Journal of Personality
and Social Psychology, 88, 658-672.
It’s For Your Own Good - 38
Author Notes
Miguel Moya, Department of Psychology; Universidad de Granada, Spain; Peter
Glick, Department of Psychology, Lawrence University, Wisconsin; Francisca Expósito,
Department of Psychology; Universidad de Granada, Spain; Soledad De Lemus,
Department of Psychology; Universidad de Granada, Spain; Joshua Hart, Department of
Psychology, Lawrence University, Wisconsin. This research was supported by Grant No.
SEJ2004-02817/PSIC from the Spanish Ministry of Education and Science (Ministerio de
Educación y Ciencia) to the first author.
Correspondence may be addressed either to Miguel Moya, Department of
Psychology, Facultad de Psicología, Campus de Cartuja, 18011, Granada, España, email:
mmoya@ugr.es, or Peter Glick, Department of Psychology, Lawrence University, PO Box
599, Appleton, Wisconsin, 54911, USA, email: glickp@lawrence.edu.
It’s For Your Own Good - 39
Figure Captions
Figure 1. Positive Responses as a Function of Relationship, Justification, and Women’s BS
Scores.
Figure 2. Positive Responses as a Function of Justification and Women’s BS Scores.
Figure 3. Positive Responses as a Function of Justification and BS Scores.
Figure 4. Perceived Discrimination as a Function of Sex of Participant and Justification.
It’s For Your Own Good - 40
0
1
2
3
4
5
6
7
Hostile Justification Protective Justification
Positive Responses
Low BS, Coworker
Opposes
Low BS, Husband
Opposes
High BS, Coworker
Opposes
High BS, Husband
Opposes
It’s For Your Own Good - 41
3
4
5
6
7
Hostile
Justification
No Justification Protective
Justification
Positive Responses
Low BS
women
High BS
women
It’s For Your Own Good - 42
1
2
3
4
5
6
7
No Justification Personalized
Protective
Group-Based
Protective
Positive Responses
Low BS
women
High BS
women
1
2
3
4
5
6
7
No Justification Personalized
Protective
Positive Responses
Low BS
men
High BS
men
It’s For Your Own Good - 43
1
2
3
4
5
6
7
No Justification Personalized
Protective
Group-Based
Protective
Perceived Discrimination
Men
Women
It’s For Your Own Good - 44
Footnotes
1
A 2 (Type of Justification) x 2 (Type of Relationship) x 2 (ASI Scale) MANOVA
revealed that condition did not affect ASI scores. The only significant effect was that
women scored higher on HS than on BS, F(1, 132) = 5.08, p < .001; all other Fs (1, 132) <
2.40, ns.
2
We typically found that women’s HS scores did not moderate their reactions to
being restricted. A truncated range on HS (women generally score well below the midpoint
of the scale; Glick et al., 2000, 2004) may limit its predictive utility. Also, even women
who agree with HS items may interpret them as referring to “other women” (e.g.,
feminists) and therefore HS may not predict reactions to restrictions aimed at themselves.
3
One “couple” was dropped because they were not an actual couple (the woman
had convinced a male friend to pose as her boyfriend to be eligible to participate).
