Content uploaded by Sherry Hamby
Author content
All content in this area was uploaded by Sherry Hamby on Nov 07, 2015
Content may be subject to copyright.
ORIGINAL ARTICLE
Size Does Matter: The Effects of Gender on Perceptions
of Dating Violence
Sherry Hamby &Amy Jackson
#Springer Science+Business Media, LLC 2010
Abstract Previous research has shown that people perceive
intimate partner violence (IPV) as more serious in cases
involving a male perpetrator and female victim versus other
gender combinations. This study is the first to explore reasons
for these differences. 181 undergraduates at a U.S. southeast-
ern college rated one of four dating violence vignettes that
varied by perpetrator and victim gender. Participants viewed
male-on-female violence as more frightening primarily
because males are stronger and bigger than female perpe-
trators. Physical differences were rated as significantly more
important causes of fear than other personality/relationship
dynamics. Because males are actually stronger and bigger
than females, it appears that gendered perceptions of violence
are based in real-world knowledge of gender differences, not
merely gender stereotypes.
Keywords Dating violence .Intimate partner violence .
Asymmetry
Introduction
One of the major debates in intimate partner violence (IPV)
research is whether violence perpetrated by males and females
is equivalent. Although a great deal of debate has addressed
the issue of whether male-perpetrated and female-perpetrated
IPVoccur at similar frequencies (Archer 2000;Hamby2009;
Straus and Gelles 1990), there has also been debate about
whether, when it occurs, male-perpetrated and female-
perpetrated physical IPV should be treated equivalently (e.g.,
Sorenson and Taylor 2005). Numerous survey studies have
established that, among the general public and college
students in the U.S. and Canada, IPV perpetrated by males
is perceived differently than violence perpetrated by females
(e.g., Cormier and Woodworth 2008; Harris and Cook 1994;
Poorman et al. 2003;SeelauandSeelau2005; Sorenson and
Thomas 2009). This study is guided by the moderate
asymmetry hypothesis, a feminist model which proposes
that there are genuine, moderately-sized gender differences in
physical IPV (Hamby 2009). This paper will extend previous
work in violence perceptions by exploring possible underly-
ing factors for gender-based differences in perceptions of
violence among undergraduates in the southeastern U.S.
The moderate asymmetry hypothesis asserts that the
percent of male and female perpetrators of physical IPV is
similar to that of other forms of physical assault and violent
crime. Extensive databases suggest that approximately 65%
to 80% of physical violence is perpetrated by males, and
approximately 20% to 35% perpetrated by females (cf. Hamby
2009 for a review). Criminologists have long known that an
imbalance of power can provide opportunities to would-
be perpetrators (Hamby 2004). The ability to physically
overwhelm a victim is one form of power, and men’sgreater
size and strength means they will more often find themselves
in situations where they could overpower a weaker victim.
Consistent with the moderate asymmetry hypothesis,
past studies have shown that male-on-female physical
violence is generally perceived more negatively than all
other gender patterns (female-on-male, male-on-male, and
female-on-female). In this respect, the public’s perceptions
of IPV are similar to those of the staff of most programs and
shelters serving IPV victims or otherwise trying to prevent
or ameliorate IPV. Although most existing research,
including the current study, has been with samples of U.S.
undergraduates, at least one study has also shown this
S. Hamby (*):A. Jackson
Department of Psychology, Sewanee, the University of the South,
Sewanee, TN 37383, USA
e-mail: sherry.hamby@sewanee.edu
A. Jackson
e-mail: ajackson@tulane.edu
Sex Roles
DOI 10.1007/s11199-010-9816-0
pattern in a sample of Canadian law enforcement officers
(Cormier and Woodworth 2008). Nonetheless, some
authors consider any differential perceptions as evidence
of gender bias and have argued for acknowledging that
female-perpetrated IPV is a much more significant social
problem than has generally been acknowledged (Cormier
and Woodworth 2008; Hamel 2007; Hamel et al. 2007;
Langhinrichsen-Rohling 2010; Straus 1993).
There are well-established differences between male-
and female-perpetrated violence, both against intimate
partners and against other targets. The best known and
best-established difference is that injury rates are several
times higher for male-perpetrated violence than for female-
perpetrated violence (Straus and Gelles 1990; Tjaden and
Thoennes 2000). Although the difference in injury rates is
large and expected, given the physical differences between
males and females, it is equally well-documented that most
physical IPV does not cause injury, especially IPV reported
in community or college samples (Straus 2005; Straus and
Gelles 1990; Tjaden and Thoennes 2000). This is also
similar to the pattern for other physical assaults (e.g.,
Finkelhor et al. 2009), most of which also do not produce
physical injuries. So the question remains whether there are
gender differences in IPV for non-injurious physical assaults.
One variable that has been proposed as an indicator of
gender differences in IPV is fear (Holtzworth-Munroe et al.
1997; Houry et al. 2008). Although most research on fear
responses to IPV has focused on women, there is some
research suggesting that women are more fearful of violent
partners than are men (Australian Bureau of Statistics 2006;
Houry et al. 2008). Some authors, however, have pointed
out that fear is gender stereotyped in U.S. culture, with
males taught to suppress expressions of fear (Dutton 2010;
Langhinrichsen-Rohling 2010). Thus, these authors suggest
that “fear”is not an equitable way to assess gender
differences in IPV. We agree that gender differences in
IPV-related fear alone do not indicate what the source of
those differences are (Hamby 2009). One purpose of this
study is to begin to explicate why men’s violence might be
more frightening than women’s violence by examining two
potential sources of fear: fear from physical differences (in
size and strength) and fear from personality/relationship
dynamics (such as controlling or unpredictable behavior).
Following the moderate asymmetry hypothesis, men’s
greater size and strength should play a more important role
in fear perceptions than personality or relationship factors.
Personality attributes and relationship behaviors vary less
between women and men (Hamby 2009; Hyde 2005).
This study will extend previous work on perceptions of
IPV in several ways. First, most previous work (Cormier
and Woodworth 2008; Harris and Cook 1994; Seelau and
Seelau 2005; Sorenson and Thomas 2009) has studied
attitudes about serious violence between married or cohab-
iting partners. Much of the controversy around gender
differences in IPV is based on reports of minor, non-
injurious violence by college students, however (Archer
2000). Dating couples have higher rates of violence than
married couples. Despite the fact that 18–24 year-olds are
only 11.7% of the population, this age group comprised
42% of IPV victims in recent years (Durose et al. 2005). It
would be worth exploring how college students evaluate
situations that are more similar to those that they are likely
to encounter—incidents of relatively minor, non-injurious
violence between dating college students.
The vignettes used in past research were limited in other
ways. Some vignettes (Harris and Cook 1994) described the
perpetrator as working in a blue-collar job, which could
influence perceptions as much as gender. In Harris and
Cook’s study, they included male-on-male violence, but
omitted female-on-female, so they did not have a fully
crossed design for perpetrator gender and victim gender.
Seelau and Seelau (2005) did add a female-on-female
vignette, but provided descriptions stating that the perpe-
trator and victim were the same size and weight in all
scenarios, regardless of gender. These descriptions do not
mirror real life, because men are usually taller, heavier, and
stronger than women (Lindle et al. 1997; Ogden et al.
2004), and in heterosexual relationships the social norm is
for men to be taller and bigger than their female partner
(Pierce 1996). Finally, some authors (e.g., Sorenson and
Thomas 2009) used a repeated measures design. Partic-
ipants who read several scenarios with varied perpetrator
and victim characteristics may try to avoid socially
undesirable ratings. In a between-subjects design, even if
participants guess or try to guess what is being manipulated
in the vignettes, they have no way of knowing how other
participants are rating their vignettes, and so these ratings
should be less subject to social desirability.
Finally, none of these previous studies explored possible
underlying causes for the differences in perceptions.
Sorenson and Thomas (2009) suggested that some victims,
that is women and gay men, are more “worthy”due to
possessing less structural power, but did not specifically
examine participants’beliefs about power or other potential
sources of differences.
Purpose and Hypotheses
This study will assess perceptions about IPV incidents with
different perpetrator gender and victim gender combinations.
1) We hypothesize that, as previous research has shown,
violence involving male perpetrators and female vic-
tims will be perceived as more severe and more the
responsibility of the perpetrator than violence by
female perpetrators and male victims.
Sex Roles
2) We also hypothesize that male-on-female violence will
be perceived more negatively than other dyads.
3) We hypothesize that participants will rate physical
differences between victim and perpetrator as more
important causes of victim fear than other personality
or relationship dynamics in scenarios involving male
perpetrators and female victims (versus female perpe-
trators and male victims).
4) We also hypothesize that this effect will be stronger for
the male-on-female than for other dyads.
5) We will explore whether the gender of the participant
affects any of these ratings.
Method
Participants
The sample was comprised of 181 undergraduates who
were recruited from a college in the southeastern United
States. The participants were 31% male and 69% female,
and 72% were underclassmen (freshmen or sophomores)
with a median age of 19 (range 18 to 22). The majority of
the participants’parents were married to one another (80%)
and college-educated (77% of mothers and 81% of fathers
had an undergraduate degree or higher). Most of the
participants (90%) had been in at least one dating
relationship. There were no differences between male and
female participants on any demographic variables, see the
“Results”for details.
Measures
Vignettes
Each participant was randomly assigned one of four
vignettes adapted from Harris and Cook (1994), but
describing a college dating relationship, not a married
couple. The vignettes presented couples who varied by
gender of perpetrator and victim. One vignette presented
the “classic”male perpetrator and female victim in a
heterosexual relationship, one depicted a female perpetrator
and male victim, one showed a male perpetrator and male
victim in a same-sex relationship, and one had a female
perpetrator and female victim. Below is the vignette
depicting male-on-female violence.
Tyler and Amber, two college students who have been
casually dating for about a month, just returned from
a movie. While at the movies, Amber spoke to her
friend Mike, which made Tyler upset. When they
returned to Tyler’s dorm, Tyler let Amber know how
he was feeling. The more Tyler talked, the more he
yelled and cursed, and then he grabbed Amber’s arm
tightly. When Amber said he was overreacting, Tyler
slapped her in the face. Finally, he stopped and said
that she had better be careful the next time she sees
Mike, or any boy for that matter.
The female-on-male violence vignette also used the
names Amber and Tyler. The other vignettes substituted the
following names: Tyler and Anthony for the same-sex male
pair and Nicole and Amber for the same-sex female dyad.
The friend was named “Tracy”in the vignettes depicting
female-on-male and female-on-female violence. Names
were chosen based on an internet search of common names
for individuals who are now college age (i.e., born in the
late 1980s or early 1990s). There were no other differences
between the vignettes except for names and pronouns.
Vignettes were randomly assigned to participants, produc-
ing cell sizes between 44 and 47 for the Perpetrator Gender
X Victim Gender interaction. Cell sizes ranged from 10 to
37 when Participant Gender was also factored in.
Attitude Questions for Vignettes
Several follow-up questions were adapted from Harris and
Cook (1994) and rated on a 7-point Likert scale, including
items on how violent the participant believed the incident to
be (endpoints “not violent”to “violent”), perpetrator respon-
sibility, victim responsibility (endpoints “not responsible”to
“very responsible”), and whether police or dorm staff should
be notified (“definitely not”to “definitely”). Participants were
also asked to describe how afraid they thought the victim
would be in the scenario they read, and then asked to rate
several possible reasons for the victim’s fear, which were
based on commonly cited factors in the research literature.
The fear questions were rated ona 7-point Likert scale ranging
from “Not at all”to “Completely.”
A principal factors analysis with a Promax rotation was
done on all questions about the vignettes. The Kaiser-
Meyer-Olkin measure of sampling adequacy was .79,
indicating sufficient sample size. A four-factor solution,
all with eigenvalues over 1.0, was produced showing
moderate correlations between the factors (.32 to .54),
except for a low correlation (.10) between Factors 2 and 4.
Together, the four factors accounted for 65% of the
variance. Loadings above .4 were retained on a factor.
Items on each factor were summed to create an index for
subsequent analyses.
Fear from Personality/Relationship Dynamics
Seven items loaded on Factor 1, all completing the stem,
“To what extent do you think [name of victim inserted from
vignette] was afraid of [name of perpetrator inserted from
Sex Roles
vignette] because…” Loadings are given in parentheses.
The items are: “[perpetrator] might do something else really
crazy one day”(.80), “[victim] feels like [perpetrator] owns
and controls [her/him]”(.74), “[perpetrator] will probably
do things like this to [victim] again”(.77). “[perpetrator] is
very unpredictable”(.56), “[perpetrator] has probably done
things like this to [victim] before”(.62), “[perpetrator]
might really hurt [victim] seriously one day”(.56), and
“[victim] is worried this is going to cause [her/him]
problems in school or with [her/his] friends”(.55). One
item, “[perpetrator] has a terrible temper,”double-loaded on
Factors 1 and 2 (.42 and .45 respectively) and was omitted.
There were no other double loadings. Coefficient alpha for
the index created with the items was .84.
Fear from Physical Differences
Two other items following the stem, “To what extent do
you think [victim] was afraid of [perpetrator] because”
loaded on this factor alone: “[perpetrator] is bigger than
[victim]”(.91) and “[perpetrator] is stronger than [victim]”
(1.00). Alpha for the scale of the combined items was .95.
Severity Index
Four items loaded on this factor: “As crimes go, how
violent was the incident?”(.49), “How afraid do you think
[name of victim] was during this incident?”(.56), “If you
had witnessed this incident in your dorm, how likely would
it have been that you would have called the police?”(.70)
and “In this case, should the dorm staff be notified?”(.65).
Alpha was .71.
Perpetrator Responsibility
Two items asking for attributions about “How responsible
was [name of perpetrator] for the incident?”(loading .86)
and “How responsible for the incident was [name of
victim]?”(−.81). The victim rating was reverse-scored to
get an index of perpetrator responsibility/lack of victim
responsibility that had a coefficient alpha of .82.
Demographics
The questionnaire ended with a number of items on
gender, age, year in school, and similar characteristics.
Procedure
Participants were recruited from undergraduate classes in
psychology and Spanish. Participation was voluntary and
confidential, and the students were eligible to enter a
drawing for a gift certificate to a local restaurant regardless
of whether they participated. A drawing was held at the end
of each class. The survey ended with a debriefing
statement, a chance to ask questions, and information on
who to contact for more information or for counseling. All
students chose to complete the survey.
Results
Gender Differences in Variables
Differences between male and female participants for the
study variables were examined at the univariate level to
provide a descriptive context for later analyses. There were
no significant gender differences on any demographic
variables. Most men (76.3%) and most women (69.6%)
were underclassmen, p>.75. Women averaged 19.6 (±.1)
years, men averaged 19.6 (±.2) years, p>.85. Approximate-
ly the same percentage of fathers had college educations
(81.5% for males, 80.8% for females, p>.90). The same
was true for mothers (72.7% for males, 79.2% for females,
p>.30). The majority of both groups had married parents
(82.4% for females, 74.5% for males, p>.20). Almost all
(88.5% of males, 90.4% of females) had been in at least one
dating relationship, p>.65.
See text below and Table 1for participant gender
differences in severity ratings, responsibility attributions,
fear from physical differences, and fear from personality/
relationship dynamics, which were incorporated into the
multivariate analyses as they pertain to Purpose #5.
Hypotheses 1 and 2: Gender Patterns in Severity Ratings
and Responsibility Attributions
A 3-way (Perpetrator Gender X Victim Gender X Partici-
pant Gender) between-subjects multivariate analysis of
variance (MANOVA) was performed on the severity index
and responsibility attributions. All three main effects were
significant, p<.001, partially supporting Hypothesis 1,
which stated that violence involving male perpetrators and
female victims will be perceived as more severe and more
the responsibility of the perpetrator than violence by female
perpetrators and male victims. No interactions were
significant at the multivariate level, all p>.05. Thus,
Hypothesis 2, that male-on-female violence would receive
the highest ratings of all dyads, was not confirmed.
For both Perpetrator Gender and Victim Gender, Hy-
pothesis 1 was confirmed for severity ratings but not
responsibility attributions. Male-perpetrated violence was
rated as more severe than female-perpetrated violence;
F(1, 172)= 24.25, p<.001. Violence against females was
rated as more severe than violence against males; F(1, 172)=
51.79, p<.001. Perpetrator gender and victim gender did not
Sex Roles
significantly influence responsibility attributions, both
p>.05. See Table 1for means and standard deviations for
these analyses.
Regarding Purpose #5, exploring differences due to
participant gender, female participants rated episodes of
dating violence as more severe and worthy of outside
intervention than male participants; F(1, 172)=14.27,
p<.001. Female participants also attributed more responsi-
bility to perpetrators and less to victims than male
participants; F(1, 172)=4.94, p< .05.
Hypotheses 3 and 4: Gender Patterns in Perceived Causes
of Fear
In order to directly test the premise that physical differences
were perceived as causes of victim fear more than
personality or relationship dynamics, a repeated measures
ANOVA was performed with Perpetrator Gender, Victim
Gender, and Participant Gender as between-subjects varia-
bles and the two fear indices (Fear from physical differ-
ences and Fear from personality/relationship dynamics) as
the repeated measure, labeled Cause of Fear.
Both Hypotheses 3 and 4 were confirmed. For the total
sample, physical differences were actually rated as less
important causes of victim fear than personality/relationship
dynamics, F(1, 172)=112.53, p<.001. But this pattern was
qualified by two-way interactions with the repeated factor,
indicating that the differences between physical and
personality/relationship ratings were higher in some cir-
cumstances. As predicted in Hypothesis 3, fear from
physical differences received higher ratings in vignettes
depicting male perpetrators, F(1, 172) = 68.67, p< .001, and
vignettes depicting female victims, F(1, 172)=40.11,
p<.001. The 3-way interaction between Perpetrator Gender,
Victim Gender, and Cause of Fear indicated that the
greatest differences in ratings of physical versus personal-
ity/relationship attributes were for male-on-female violence
in comparison to other dyads, F(1, 172)=6.03, p< .05,
supporting Hypothesis 4. See Fig. 1.
Regarding our final purpose, to explore the effects of
participant gender, we also found that the difference in
physical versus personality/relationship ratings were higher
for female versus male participants, F(1, 172) = 8.45, p<.01.
A 3-way interaction indicated this effect was particularly
strong when female participants rated male perpetrators, F
(1, 172)=6.62, p<.05.
Between-subjects tests also showed that male perpetra-
tors and female victims received overall higher ratings than
female perpetrators and male victims, and that male-on-
female scenarios had higher ratings than other vignettes.
For Perpetrator Gender, F(1,172)= 84.96, p< .001; for
Victim Gender, F(1,172)=46.73, p< .001; and for Perpetra-
tor Gender X Victim Gender, F(1,172)= 10.44, p< .01.
Further analyses indicated these patterns were evident for
both physical and personality/relationship variables for
perpetrators, but only for physical differences for victims.
No other between-subjects effect was significant, p> .10.
See Table 1.
1
2
3
4
5
6
Male-on-
female
Male-on-
male
Female-on-
female
Female-on-
male
Physical differences
Personality/relationship
dynamics
Fig. 1 Physical differences were perceived to have a stronger impact
on fear for male-on-female violence in comparison to other dyads. In
contrast, personality/relationship dynamics were perceived more
similarly across all relationship types.
Table 1 Means (and standard deviations) for vignette ratings for full sample and by perpetrator gender, victim gender, and participant gender.
Item Total (n=181) Perpetrator Victim Participant
Male
(n=88)
Female
(n=93)
Male
(n=89)
Female
(n=92)
Male
(n=55)
Female
(n=125)
Severity index 4.39 (1.2) 4.79 (1.2) 4.00 (1.1)*** 3.88 (1.2) 4.91 (1.0)*** 4.07 (1.2) 4.54 (1.2)***
Perpetrator responsibility index 6.04 (1.3) 6.19 (1.3) 5.89 (1.2) 5.84 (1.3) 6.25 (1.2) 6.18 (1.1) 5.73 (1.5)*
Fear from physical differences 4.06 (1.8) 5.22 (1.0) 2.96 (1.7)*** 3.27 (1.8) 4.89 (1.4)*** 4.35 (1.5) 3.96 (1.9)
Fear from personality/relationship 5.26 (.9) 5.46 (.8) 5.06 (1.0)*** 5.39 (.8) 5.12 (1.0) 5.00 (1.0) 5.38 (.9)
Participants rated one of four vignettes which experimentally manipulated perpetrator and victim gender. Ratings were made on a 7-point Likert
scale; higher ratings indicated higher perceptions of severity, responsibility, or causes of fear. See text and Fig. 1for interaction effects.
*p<.05; **p< .01; ***p< .001. n= 181
Sex Roles
Discussion
The results showed that, in this sample from the southeast-
ern United States, gender of the perpetrator and of the
victim can alter perceptions about the severity, responsibil-
ity, and causes of fear for physical violence in dating
relationships. Students perceived that male perpetrators
were more frightening than female perpetrators, and that
the reasons for differences in fear were due more to
physical differences between male perpetrators and female
victims than because of personality or relationship factors.
These findings our consistent with the moderate asymmetry
hypothesis of gender differences in IPV (Hamby 2009).
Our finding that violence involving male perpetrators and
female victims is perceived as more severe is consistent
with other literature using similar methodologies, and
shows that these patterns are consistent across scenarios
using married couples, for example as in Harris and Cook
(1994), as well as dating couples as found here. Taken
together, the findings across several vignette studies
(Cormier and Woodworth 2008; Harris and Cook 1994;
Seelau and Seelau 2005; Sorenson and Thomas 2009)
suggest that the effect of gender on perceptions of severity
is relatively stable for relationships of varying length,
commitment, and violence severity.
Many authors have described differences in perceptions
based on the gender of the actors as due to “gender
stereotypes”(Cormier and Woodworth 2008; Seelau and
Seelau 2005)or“social norms”(Sorenson and Taylor
2005). Although these terms are not inappropriate in and of
themselves, they can suggest that such ratings are biased in
some ways. Indeed, some authors seem to interpret a lack
of differences based on the gender of the perpetrator or
victim as evidence of objectivity, as when Seelau and
Seelau describe their lack of differences found for ratings of
general aggressiveness as a function of perpetrator gender
(2005). In real life, however, men are more generally
aggressive than women (Hyde 2005), and so rather than
representing objectivity, one could argue that such ratings
indicate a pro-male bias. Some authors (Cormier and
Woodworth 2008; Dutton 2010; Hamel et al. 2007) even
more explicitly attribute any gender-based perceptions as
signs of bias, despite the real-world differences between
male- and female-perpetrated violence.
There are, in fact, objective differences between the
genders. Although many measures of social equality have
improved somewhat in the last several decades (Institute for
Wom e n ’s Policy Research 2009), physical differences
between the genders remain. Males are approximately
5 in. taller and 27 lb heavier than women (Ogden et al.
2004). Interestingly, despite the fact that both males and
females are taller and heavier than they were in the past, the
height and weight differences between males and females
have stayed almost exactly the same over the last 40 years
or so. The average weight of both men and women has
increased by more than 24 lbs since 1960, but the relative
difference was 26.1 lbs in 1960 and was 26.7 lbs in 2002
(Ogden et al. 2004). Men are also substantially stronger
than women (Lindle et al. 1997). Further, in the United
States, as in many other countries, social norms promote a
male-taller norm for romantic pairings (Pierce 1996).
Finally, males are more violent than females across the
lifespan (Hamby 2009; Hyde 2005; Tjaden and Thoennes
2000). These differences all contribute to moderate gender
asymmetry in patterns of perpetrating physical violence,
both against intimate partners and against others.
It certainly seems possible that evolutionary, political, or
social trends might change the relative physical differences
between men and women or other behaviors that will affect
the relative severity of male-on-female versus female-on-
male violence. For example, although we are aware of no
formal research on the topic, there has been some recent
media attention suggesting that the male-taller norm may be
relaxing (Hall 2009). Until physical parity in size and
strength becomes the rule rather than the exception,
however, male-on-female physical violence will always,
on average, be more severe and more frightening than other
gender combinations. The participants’ratings are accu-
rately reflecting these realities.
Same-Sex and Opposite-Sex Relationships
As with previous research (e.g., Seelau and Seelau 2005),
we did not find many differences suggesting that violence
in same-sex relationships is perceived very differently than
violence in heterosexual relationships. The above-noted
findings that size and strength differences are smaller in
cases of violence between same-sex partners versus male-
on-female partners is consistent with actual gender differ-
ences in size and strength and also with existing data
indicating that female victims with male partners are at
greater risk of injury than female victims with female
partners (Tjaden and Thoennes 2000).
Limitations
Some limitations of the study include the homogeneous nature
of the participants. All were college-aged, most were
Caucasian, and most came from intact families and privileged
socioeconomic backgrounds. It would be worthy to replicate
this study with a more varied participant group, particularly in
a different part of the world. The focus of this study was on an
incident that included physical dating violence, and percep-
tions may vary for other types of IPV, including psychological
and sexual aggression. Physical differences in size and
strength may be less likely to influence perceptions of
Sex Roles
psychological aggression when it occurs in the absence of
physical violence or threats of violence. Fewer gender
differences have been found for psychological than physical
aggression in all types of relationships (Hyde 2005), while
more gender differences have been found for sexual
aggression (Hamby 2009). Further, although the experimen-
tal design allowed us to control factors other than gender, it
also meant that we focused on a single type of violence in a
particular context. Despite the limitations of this study, it
produced results that shed further light on the role of gender
in intimate partner violence.
Implications for Future Research
Much more needs to be done to explore the differential
impact of violence by men and women. Further studies
could explore whether impact, or perceptions of impact,
vary for other types of IPV such as psychological
aggression and sexual assault. Other possible causes of
fear could also be explored. For example, Stark (2010) has
suggested that gender differences in fear are due to
differential patterns of coercive control rather than gender
stereotypes. Other types of differential impact could also be
explored—for example, is violence by men more likely to
interfere with work, income, school, or other relationships?
We need to know much more about the differences between
men’s and women’s violence.
Conclusion
Seen in the larger context of criminological research on
physical assault, male-on-female dating violence follows a
common pattern in which assailants take advantage of
greater size or strength (Browne and Finkelhor 1986;
Hamby 2004). From bullies to muggers, most assailants
choose victims who are physically weaker than themselves,
and violence involving such power differentials warrants
greater attention and intervention from law enforcement
and human service professionals. The well-established
findings that male-on-female violence is perceived more
negatively than other forms of IPV appears to be primarily
due to the equally well-established physical differences
between males and females.
References
Archer, J. (2000). Sex differences in aggression between heterosexual
partners: A meta-analytic review. Psychological Bulletin, 126,
651–680.
Australian Bureau of Statistics. (2006). Personal safety survey:
Australia (Reissue). Sydney: Author.
Browne, A., & Finkelhor, D. (1986). Impact of child sexual abuse: A
review of the research. Psychological Bulletin, 99,66–77.
Cormier, N. S., & Woodworth, M. T. (2008). Do you see what I see?
The influence of gender stereotypes on student and Royal
Canadian Mounted Police perceptions of violent same-sex and
opposite-sex relationships. Journal of Aggression, Maltreatment,
& Trauma, 17,1–26.
Durose, M., Harlow, C. W., Langan, P. A., Motivans, M., Rantala,
R. R., & Smith, E. L. (2005). Family violence statistics including
statistics on strangers and acquaintances. Washington, DC:
Bureau of Justice Statistics.
Dutton, D. G. (2010). The gender paradigm and the architecture of
antiscience. Partner Abuse, 1,5–25.
Finkelhor, D., Turner, H., Ormrod, R., & Hamby, S. (2009). Violence,
abuse and crime exposure in a national sample of children and
youth. Pediatrics, 124, 1411–1423.
Hall, K. (2009, June 23). Famous couples with taller women: Is height
just a number? Retrieved from http://www.huffingtonpost.com/
2009/06/14/famous-couples-with-talle_n_214948.html.
Hamby, S. L. (2004). The spectrum of victimization and the implications
for health. In K. Kendall-Tackett (Ed.), Health consequences of
abuse in the family: A clinical guide for practice (pp. 7–27).
Washington, DC: American Psychological Association.
Hamby, S. L. (2009). The gender debate on intimate partner violence:
Solutions and dead ends. Psychological Trauma, 1,24–34.
Hamel, J. (2007). Toward a gender-inclusive conception of intimate
partner violence research and theory: Part 1, Traditional
perspectives. International Journal of Men’s Health, 6,36–53.
Hamel, J., Desmarais, S., & Nicholls, T. (2007). Perceptions of
motives in intimate partner violence: Expressive versus coercive
violence. Violence and Victims, 22, 563–576.
Harris, R. J., & Cook, C. A. (1994). Attributions about spouse abuse:
It matters who the batterers and victims are. Sex Roles, 30, 553–
565.
Holtzworth-Munroe, A., Smutzler, N., & Bates, L. (1997). A brief
review of the research on husband violence, Part III: Sociodemo-
graphic factors, relationship factors, and differing consequences
of husband and wife violence. Aggression and Violent Behavior,
2, 285–307.
Houry, D., Rhodes, K. V., Kemball, R. S., Click, L., Cerulli, C.,
McNutt, L., et al. (2008). Differences in female and male victims
and perpetrators of partner violence with respect to WEB scores.
Journal of Interpersonal Violence, 23, 1041–1055.
Hyde, J. S. (2005). The gender similarities hypothesis. The American
Psychologist, 60, 581–592.
Institute for Women’s Policy Research. (2009). The gender wage gap:
2008. Washington, DC: Author.
Langhinrichsen-Rohling, J. (2010). Controversies involving gender
and intimate partner violence in the United States. Sex Roles, 62,
179–193.
Lindle, R., Metter, E., Lynch, N., Fleg, J., Fozard, J., Tobin, J., et al.
(1997). Age and gender comparisons of muscle strength in 654
women and men aged 20–93 yr. Journal of Applied Physiology,
83, 1581–1587.
Ogden, C. L., Fryar, C. D., Carroll, M. D., & Flegal, K. M. (2004).
Mean body weight, height, and body mass index, United States
1960–2002. Atlanta: National Center for Health Statistics,
Centers for Disease Control and Prevention.
Pierce, C. A. (1996). Body height and romantic attraction: A meta-
analytic test of the male-taller norm. Social Behavior and
Personality, 24, 143–150.
Poorman, P., Seelau, E. P., & Seelau, S. M. (2003). Perceptions of
domestic abuse in same-sex relationships and implications for
criminal justice and mental health responses. Violence and
Victims, 18, 659–669.
Seelau, S. M., & Seelau, E. P. (2005). Gender-role stereotypes and
perceptions of heterosexual, gay and lesbian domestic violence.
Journal of Family Violence, 20, 363–371.
Sex Roles
Sorenson, S. B., & Taylor, C.A. (2005). Female aggression towards male
intimate partners: An examination of social norms in a community-
based sample. Psychology of Women Quarterly, 29,78–96.
Sorenson, S. B., & Thomas, K. A. (2009). Views of intimate partner
violence in same-and-opposite-sex relationships. Journal of
Marriage and Family, 71, 337–352.
Stark, E. (2010). Do violent acts equal abuse? Resolving the gender
parity/asymmetry dilemma. Sex Roles, 62, 201–211.
Straus, M. A. (1993). Physical assaults by wives: A major social
problem. In R. Gelles & D. Loseke (Eds.), Current controversies
on family violence (pp. 67–87). Newbury Park: Sage.
Straus, M. A. (2005). Gender and partner violence in world
perspective: Some results from the International Dating Violence
Study. Paper presented at the 9th International Family Violence
Research Conference, Portsmouth, NH.
Straus, M. A., & Gelles, R. J. (1990). Physical violence in American
families: Risk factors and adaptations to violence in 8, 145
families. New Brunswick: Transaction.
Tjaden, P., & Thoennes, N. (2000). Extent, nature, and consequences
of intimate partner violence: Findings from the national violence
against women survey. Washington, DC: National Institutes of
Justice.
Sex Roles
