Content uploaded by Virginia Sánchez
Author content
All content in this area was uploaded by Virginia Sánchez on Nov 04, 2014
Content may be subject to copyright.
PLEASE SCROLL DOWN FOR ARTICLE
This article was downloaded by:
[Menesini, Ersilia]
On:
20 February 2011
Access details:
Access Details: [subscription number 933728305]
Publisher
Psychology Press
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-
41 Mortimer Street, London W1T 3JH, UK
European Journal of Developmental Psychology
Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713684600
Reciprocal involvement in adolescent dating aggression: An Italian-
Spanish study
Ersilia Menesini
a
; Annalaura Nocentini
a
; Francisco Javier Ortega-Rivera
b
; Virginia Sanchez
b
; Rosario
Ortega
c
a
Department of Psychology, University of Florence, Florence, Italy
b
Department of Developmental
and Educational Psychology, University of Seville, Seville, Spain
c
Department of Psychology,
University of Córdoba, Córdoba, Spain
First published on: 18 February 2011
To cite this Article Menesini, Ersilia , Nocentini, Annalaura , Ortega-Rivera, Francisco Javier , Sanchez, Virginia and
Ortega, Rosario(2011) 'Reciprocal involvement in adolescent dating aggression: An Italian-Spanish study', European
Journal of Developmental Psychology,, First published on: 18 February 2011 (iFirst)
To link to this Article: DOI: 10.1080/17405629.2010.549011
URL: http://dx.doi.org/10.1080/17405629.2010.549011
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
Reciprocal involvement in adolescent dating aggression:
An Italian–Spanish study
Ersilia Menesini
1
, Annalaura Nocentini
1
,
Francisco Javier Ortega-Rivera
2
, Virginia Sanchez
2
, and
Rosario Ortega
3
1
Department of Psychology, University of Florence, Florence, Italy
2
Department of Developme ntal and Educational Psychology, University of
Seville, Seville, Spain
3
Department of Psychology, University of Co
´
rdoba, Co
´
rdoba, Spain
This study aimed to: (1) analyse the association of different patterns of
reciprocal involvement in dating aggression (reciprocal psychological and
physical aggression; reciprocal psychological aggression; non-aggressive
individuals) with different qualities of romantic relationship; and (2)
compare results in two European countries, Italy and Spain. Participants
were 304 adolescents (141 Italy, 163 Spain) with a current dating relationship.
Results in both countries showed that adolescents involved in reciprocal
psychological and physical aggression are more likely to have higher levels of
couple conflict and power imbalance as compared to the reciprocal
psychological aggressive group and to the non-aggressive individuals.
Besides, adolescents involved in reciprocal psychological aggression are more
likely to have lower levels of support as compared to the non-aggressive
individuals. Discussion is focused on different patterns of reciprocal
involvement differentiated in relation to level of aggression, conflict and
power imbalance.
Keywords: Dating aggression; Reciprocal involvement; Romantic relationship
quality; Adolescence; Cross-cultural comparison.
Correspondence should be addressed to Ersilia Menesini, Department of Psychology,
University of Florence, Via di San Salvi, 12, Complesso di San Salvi Padiglione 26, I-50135,
Florence, Italy. E-mail: menesini@psico.unifi.it
This research was financed by the Italian Ministry of Education and Research (Bilateral
special action project AIIS0518EE/2006) and by the Spanish Ministry of Science and
Innovation (HI2005-0452).
EUROPEAN JOURNAL OF DEVELOPMENTAL PSYCHOLOGY
0000, 00 (00), 000 – 000, iFirst article
Ó 2011 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business
http://www.psypress.com/edp DOI: 10.1080/17405629.2010.549011
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
INTRODUCTION
Descriptive studies on dating aggression have unde rlined a pattern of
mutual violence involving psychological and physica l aggression from both
partners (Archer, 2000; Capaldi & Crosby, 1997; Frieze, 2005). Especially in
adolescence both male and female partners were frequently found to be
involved as perpetrators and victims, suggesting a reciprocal involvement in
dating aggression (Capaldi & Crosby, 1997; Cascardi & Vivian, 1995; Gray
& Foshee, 1997; Johnson, 1995; O’Leary & Slep, 2003). Studies conducted
on adolescents in the USA reported a frequency range of 53% to 72% of
mutually aggressive dating relationships (Gray & Foshee, 1997; Henton,
Cate, Koval, Lloyd, & Christopher, 1983). A similar figure was also found in
Italy (Menesini & Nocentini, 2008). Mutual violence more often implies
mild forms of aggression, however severe mutual aggression can also be
found occasionally (Johnson, 1995; Olson, 2002; Williams & Frieze, 2005).
Most studies on mutual aggression have used a group-based approach in
order to differentiate mutual from one-sided pattern couples (Gray &
Foshee, 1997; Henton et al., 1983). Using the same group-based approach,
the present study intended to focus on patterns of mutual involvement in
aggressive behaviours, and to further investigate if diffe rent degrees of
severity in the reciprocal patterns of aggression (psychological vs.
psychological and physical) can be associated with different qualities of
romantic relationship. Following the escalation conflict theory we
hypothesized that the level of conflict severity can escalate when it passes
from verbal and psychological attacks to physical aggression. On this basis
two main groups can be identified a priori within the mutual couple
aggression as they represent distinct patterns of involvement: one group,
where the partners are characterized by a reciprocal involvement in
psychological dating aggression (RPA), and another group where the
partners are characterized by a reciprocal involvement in both psychological
and physical dating aggression (RPPA). The aim of the study was to analyse
whether these theoretical patterns of reciprocal involv ement in dating
aggression were empirically associated with different qualities of romantic
relationship in terms of conflict, imbalance of power and support. In
addition, data from Italy and Spain were to be compared in order to find
commonalities and differences. Literature has stressed the relevance of
understanding the moderating role of culture on adolescent romantic
relationships and dating aggression (Seiffge -Krenke & Connolly, 2010),
since youth from different cultures are more and more connected in
relationships. Italy and Spain share several cultural features and have
similar figures for domestic violence and violence against women : however,
no cross-cultural comparison between these two countries has yet been
conducted in relation to dating aggression. The aim of this cross-cultural
2
MENESINI ET AL.
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
comparison was to understand whether two similar European countries
shared the same features and mechanisms leading from different qualities of
romantic relationships (conflict, imbalance of power and support) to
different or similar patterns of dating aggression.
Reciprocal dating aggression
The theoretical model through which mutual couple violence or situational
couple violence is studied derives from the work of family violence
researchers (see Straus & Gelles, 1990). This model refers to the conflict
between the two partners where physical aggressive acts are tactics used in
response to a conflict, as opposed to the coercive approach in unilateral
violent relationships. Situational couple violence occasionally takes place
when conflict gets out of hand: it is not a general means of controlling the
other partner but it is the result of a temporary attempt to establish control
during a conflict (Johnson, 1995; Kelly & Johnson, 2008). This pattern of
aggression is characterized by gender symmetry, defined as equivalent rates
of partner violence by males and females (Archer, 2000; Frieze, 2005) and is
distinct from the coercive approach, where one partner, usually the male, is
the perpetrator and the other the victim and where physical aggression is a
means of maintaining a general control over the partner.
The dynamics through which conflict, reciprocity, imbalance of power
and different levels of aggression relate to each other is significant in
explaining the process of partner conflict escalating to violence. According
to a dynamic developmental system approach (Capaldi & Kim, 2007),
interactions between partners can provoke or reinforce aggressive acts
within the dyad, particularly one individual’s behaviour may elicit similar
acts from the partner. For example, starting from an argument, verbal or
physical aggression may be used to respond to a verbal conflict or
opposition. Higher levels of conflict can easily escalate towards higher levels
of dating aggression, which in turn contrib ute to reinforce and expand the
spiral of conflict (Straus & Gelles, 1990). From these models we surmise that
psychological and verbal aggression represent a first manifestation of the
conflict, which may escalate towards more violent acts, particularly when
one partner or both are lacking in communication skills. In fact, several
studies have shown that psychologically aggressive behaviours precede and
predict the development of physically aggressive behaviours, supporting the
hypothesized progression from psychological to physical aggression in
young couples (O’Leary & Woodin, 2009). The desire to maintain or to
(re)establish individual control within the conflictual exchange, a process
involving both males and females, can be related to this progression
(Connolly et al., 2010; Frieze, 2005). Olson (2002), analysing patterns of
communication in escalating conflictual interactions, made a distinction
CROSS-CULTURAL DATING AGGRESSION 3
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
between aggressive and violent relationships. The aggressive couple has a
symmetrical relationship in which both parties share power and are equally
prone to use low-to-medium aggression, including verbal and non-contact
physical acts. The violent couple, too, has a symmetrical relationship in
terms of power and violence, but in this case the violence includes phy sical
acts and severe verbal aggression, and both partners act violently in order to
maintain or (re)establish individual control.
Moving from these considerations, the present study aimed to analyse if
and how different theoretically defined patterns of reciprocal involvement
in dating aggression (psychological vs. psychological and physical) are
associated with different qualities of dating relationship, i.e., level of
conflict, imbalance of power and degree of support. Although the
escalation of conflict can be related to a causal and developmental model
(from verbal or psychological conflict to physical aggression), in the
present study we considered a cross-sectional perspective where the
sequential steps of the conflict dynamics were represented by a distinct
pattern of relationship between the two partners. We assumed that RPPA
and RPA groups differentiate from the couples where partners are not
involved (NI) in any type of partner violence by their more compromised
profile characterized by higher levels of conflict and lower levels of
support. Furthermore, the RPPA group might show a higher level of
conflict and imbalance of power along with a lower level of support
compared to RPA. Given the relevance of gender in this research field, all
the analyses were controlled for possible gender effects. However, as
gender symmetry was found in terms of frequency of behaviours (Archer,
2000) and gender invariance was found in terms of dating aggression
processes involving conflict and imbalance of power (Connolly et al.,
2010), we assumed that gender does not affect the relations between
quality and involvement in different recipr ocal profiles.
Cross-cultural considerations
Most studies on dating aggression have been conducted in the USA,
Canada, the UK and New Zealand (see Archer, 2006) and only recently has
research extended to Italy and Spain. Although some studies have already
been published in both countries (Menesini & Nocentini, 2008, 2009;
Mun
˜
oz-Rivas, Gran
˜
a, O’Leary, & Gonza
´
lez, 2007a, 2007b), none presents a
direct comparison between the two countries. A previous study showe d the
reciprocal involvement of partners in both roles of perpetrator and victim in
Italy (Menesini & Nocentini, 2008), but no specific investigation has yet
been conducted on this profile among Spa nish adolescents.
On a general level, Italy and Spain are similar nations. They are
neighbouring Western European nations whose cultures, languages, and
4
MENESINI ET AL.
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
social identities have much in common. Italians and Spaniards have a
similar approach to life and share the same types of values in business,
education, and in family and religious contexts. In relation to partner
violence, the high levels reported in both countries have recently drawn
political attention to the problems of family and gender-based violence. A
large national study on women carried out in Italy reported that 14.3% of
participants had been victims of physical or sexual assault by their own
partner or ex-partner at least once during their life (Italian Institute of
Statistics, 2006). In Spain, a similar study showed a figure of 9.6% of women
reporting partner violence (Instituto de la Mujer, 2006). Focusing on
adolescence and young adulthood, physical dating aggression in Spain
showed a higher involvement than in American samples (Mun
˜
oz-Rivas
et al., 2007b); percentages of involvement in Italy are similar to those
obtained in North America for physical dating aggression but higher for
psychological aggression (Connolly et al., 2010). From these considerations,
we expected to find a higher involvement in dating aggression in the Spanish
sample as compared to the Italian one.
In relation to predictors of dating aggression, a study comparing Italian
and Canadian youth showed that con flict was a significant factor in both
groups, whereas the pathway from power imbalance to dating aggression
was significant just for Italy (Connolly et al., 2010). This result underlines
the fact that in Italy the perception of imbalance of power in the dyad is a
possible cause for fighting. Are Italian and Spanish adolescents similar in
relation to mechanisms leading to reciprocal dating aggression? Given the
cultural similarity between Italy and Spain, we expected to find similar
structural paths in the two countries.
The present study
The present study aimed to: (1) define groups of participants characterized
by different profiles of involvement in aggressive acts—reciprocal involve-
ment in physical and psychological aggression (RPPA), reciprocal involve-
ment in psychological aggression only (RPA) and not involved (NI) and
evaluate gender and country differences; and (2) analyse whether the
multivariate effects of predictors (gender, conflict, imbalance of power and
support) on different dating aggression profiles was moderated by country.
METHOD
Sample
Adolescents (N ¼ 588) from two European cities (Florence, Central Italy)
and Seville (Southern Spain) participated in the study. Since the present
CROSS-CULTURAL DATING AGGRESSION 5
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
study focused only on students who had a current romantic relationship, 284
students were excluded from the analysis as they were neither having a
romantic relationship nor had had one in the past. The final sample
consisted of 304 adolescents (163 Spain; 141 Italy). Sa mple characteristics
are shown in Table 1. The sample was representative of the school
distribution in Italy and Spain.
TABLE 1
Sample descriptive statistics
Italy Spain Diff.
Sex Male 52 (36.9%) 58 (35.6%) ns
Female 89 (63.1%) 105 (64.4%)
Age (Mean and SD) 17.29 (0.982) 17.14 (1.094) ns
Parent education Elementary and
middle school
39 (27.8%) 124 (75.2%) w
2
(2, 488) ¼
159.497***
High-school
degree
69 (48.9%) 16 (10.2%)
University
degree
33 (23.1%) 123 (14.5%)
Ethnicity Italy/Spain 136 (96.6%) 160 (98.6%) ns
Others 5 (3.4%) 3 (0.4%)
Mean relationship
length
(Mean and SD)
45.31 (46.26) 56.01 (53.51) ns
Quality of romantic
relationship
(Mean and SD)
Conflict 2.11 (0.71) 2.30 (0.77) F(1, 298) ¼
5.053*
Imbalance
of power
2.01 (1.03) 1.71 (0.81) F(1, 295) ¼
8.434**
Support 3.68 (0.88) 4.22 (0.66) F(1, 298) ¼
36.995***
Dating aggression
(Mean and SD)
Psychological
dating
aggression
perpetrated
0.46 (0.62) 0.49 (0.56) ns
Psychological
dating
aggression
received
0.36 (0.59) 0.42 (0.56) ns
Physical dating
aggression
perpetrated
0.12 (0.28) 0.24 (0.43) F(1, 292)
¼ 8.006**
Physical dating
aggression
received
0.11 (0.27) 0.20 (0.34) F(1, 292)
¼ 5.593*
6 MENESINI ET AL.
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
Measures
Quality of dating relationships. Dating relationship quality was mea-
sured by the Network Relationships Inventory (NRI; Furman & Buhrme-
ster, 1992). This measure comprises 17 items, which assess three subscales on
a 5-point Likert scale ranging from nev er true to always true: (1) nine items
assess ‘‘support’’ defined as communication (I tell my boy/girlfriend things I
would not want others to know) and commitment (I’m sure this relationship
will continue in the future); (2) six items were related to conflict in the dyad
(I get upset with my boy/girlfriend); and (3) two items were related to
imbalance of power in the dyad (How often does someone tend to be bossy
in this relationship?).
Dating aggression. Dating aggression was assessed using four scales
measuring psychological and physical aggression perpetrated and received.
Although physical aggression is often reported as the main index of dating
aggression, the relevance of considering other verbal and psychological acts
was underlined by several authors on account of their interrelated nature
(Capaldi & Crosby, 1997; O’Leary & Slep, 2003). Following the definition
by Straus (1979) and Capaldi and Crosby (1997), we decided to include
verbal aggression, offence and threatening in the category of psychological
aggression. The first two scales comprised nine items for perpetrated acts
and nine items for received acts, derive d from the Physical Violence Scale of
Conflict Tactics Scale (Nocentini et al., in press; Straus, 1979), i.e., pushing,
grabbing, or shoving; slapping, kicking , or biting. The second two scale s
were derive d from the Relational Aggression Scale (Crick, 1995) and
consisted of three items focused on psychological aggression perpetrated
and three items on psychological aggression received (spreading rumours or
telling mean lies to make one’s boy/girlfriend unpopular; telling one’s boy/
girlfriend she/he won’t be liked anymore unless she/he does what she/he is
being told to do; teasing). The items of the four scales were rated on a 5-
point Likert scale, ranging from never to always.
Plan of analyses
The analysis consisted of three steps. As this study was based on a cross-
cultural comparison, all the scales used had to have measurement
equivalence across groups in order to obtain meaningful differences (Collins,
Raju, & Edwards, 2000). As reported by Collins and colleagues (2000), one
of the best methods to determine scale and measure accuracy is to analyse
the equivalence of the construct among different sociodemographic groups.
If the structure is confirmed across groups—specifically the metric and
scalar invariance—the latent factor and the measure can be generalized.
CROSS-CULTURAL DATING AGGRESSION 7
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
Therefore, as a preliminary step, we conducted a multiple-group con-
firmatory factor analysis across country for the variables considered (a series
of nested models adding more stringent requirements progressively were
tested; Vandenberg & Lance, 2000). In the second phase we distinguished
three different profiles of reciprocal involvement in dating aggressive acts.
Finally, we conducted a multiple-group multinomial logistic regression
across country, considering the three profiles of reciprocal involvement in
aggressive behaviours as dependent variables and gender and quality of
romantic relationships as independent variables.
All these analyses were conducted using Mplus 4.0 (Muthe
´
n & Muthe
´
n,
2006). The models were evaluated by means of the following overall indices:
the chi-square statistic, the root-mean-square error of approximation
(RMSEA), the comparative fit index (CFI), the Akaike information criteria
and the Bayesian information criteria. Recommended cut-off points for
these measures are: for RMSEA the cut-off is .08 (Brown & Cudek, 1993)
and for CFI it is .90 (Bollen, 1989). AIC and BIC models that fit better have
smaller values on these statistics.
RESULTS
Preliminary analysis
With regard to romant ic relationships quality, we tested a model with three
correlated latent factors: support, conflict and imbalance of power. Final
results showed that, deleting one item, full metric and full scalar invariance
across country was demonstrated (see Table 2). All the alpha coefficients
were acceptable for both countries (imbalance of power: Italy ¼ .79; Spain ¼
.68; conflict: Italy ¼ .81; Spain ¼ .86; support: Italy ¼ .90; Spain ¼ .84).
With regard to dating aggression, two models (one for perpetration and
one for victimization) with two correlated latent factors (physical and
psychological dating aggression) were tested (see Table 2). A full metric and
scalar invariance was found for the perpetrated behaviours (Model 2).
Latent factors correlations were .66 and .80 in Italy and Spain, respectively.
Acceptable alpha coefficients for physical dating aggression perpetrated
were found (Italy ¼ .76; Spain ¼ .73), but they were poor for psychological
dating aggression perpetrated (Italy ¼ .58; Spain ¼ .57). Since these scales
are composed of only three items, and the average inter-item tetrachoric
correlations are .48 and .32, respectively, we retained these two scales in the
following analysis. A full metric and scalar invariance was found for the
received behaviours (Model 2). Correlations between latent factors were .78
and .71 in Italy and Spain, respectively. All the alpha coefficients were
acceptable (physical and psychologi cal dating aggression received were:
Italy ¼ .84; Spain ¼ .67; Italy ¼ .63; Spain ¼ .63, respectively).
8
MENESINI ET AL.
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
TABLE 2
Fit indices for multiple-group confirmatory factor analysis across country on romantic relationship quality and on dating aggression
w
2
df p CFI RMSEA AIC BIC Dw
2
Ddf P
Romantic relationship quality
1. Configural invariance 353.957 230 .000 .94 .062 (.049–.065)
2. Metric invariance 385.250 244 .000 .93 .065 (.052–.077)
3. Configural invariance
a
329.658 201 .000 .93 .068 (.054–.081) 11224.19 11598.21
4. Metric invariance 350.953 214 .000 .93 .068 (.055–.080) 11219.51 11546.32
5. Scalar invariance 371.778 226 .000 .92 .068 (.055–.080) 11216.31 11499.55
2 vs. 1 31.29 14 .01
4 vs. 3 21.295 13 .07
5 vs. 4 20.825 12 .05
Psychological and physical dating aggression perpetrated
1. Configural invariance 46.174 22 .002 .93 .088
2. Metric and scalar invariance
b
50.904 24 .001 .92 .088
2 vs. 1 9.042 4 .060
Psychological and physical dating aggression received
1.Configural invariance 20.398 19 .371 .99 .023
2. Metric and scalar invariance 21.660 19 .301 .98 .031
2 vs. 1 4.993 3 .172
Note: CFI ¼ Comparative Fit Index; RMSEA ¼ Root Mean Square Error of Approximation; AIC ¼ Akaike information criteria; BIC ¼ Bayesian
information criteria; Dw
2
¼ differences between the two nested models w
2
statistics; Ddf ¼ differences between degree of freedom of the two nested models.
a
The item ‘‘I go places and do enjoyable things with my boyfriend/girlfriend’’ from the Support scale was deleted given that it showed non-invariance
across groups.
b
The items of Dating Aggression scales are not normally distributed, thus we dichotomized them in presence/absence and we analysed the
data using WLSMV estimator for categorical data. Using this procedure, metric and scalar invariance need to be tested at the same step and AIC and
BIC are missing.
9
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
Prevalence of different reciprocal profiles
In order to create different profiles of reciprocal aggressive acts we
proceeded as follows: first, we created the dichotomous prevalence for each
of the four scales (presence ¼ ad olescents involved in at least one behaviour).
Second, we created the dichotomous reciprocal index for each psychological
and physical scale: we defined presence of reciprocal dating aggression when
adolescents reported a presence of dating aggression perpetrated and
received. Third, we created three groups of participants: (1) RPPA are those
participants who are reciprocally involved as perpetrators and victims in
physical and in psychological aggressive behaviours; (2) RPA are those who
are reciprocally involved as perpetrators and victims only in psychological
aggressive behaviours; and (3) NI are those who are not involved in
reciprocal dating aggressive relationships. Students with missing data in
some variables were excluded. Overall, the following analyses involved 273
adolescents: 130 Italians (86 females) and 143 Spanish (94 females).
Results showed that out of the total sample, 46 (16.9%) students were
classified as RPPA, 97 (35.5%) as RPA, and 130 (47.6%) as NI. In Spain,
the percentages were: 20.3% for RPPA (22% males; 19% females), 37.8%
for RPA (39% males; 37% females) and 42% for NI (39% males; 44%
females): gender differences were not found in this sample, w
2
(2,
143) ¼ 0.373; p ¼ .830. In Italy the corresponding percentages were: 13.1%
for RPPA (20% males; 9% females), 33.1% for RPA (19% males; 41%
females) and 53.8% for NI (61% males; 50% females): the test of chi-square
showed slight significant gender differences, w
2
(2, 130) ¼ 7.928; p ¼ .02, but
no standardized residuals were significant. Overall, no significant differences
were found across country, w
2
(2, 273) ¼ 4.538; p ¼ .103.
Modelling the association between relationship quality and
different profiles of dating aggression
A multiple-group multinomial logistic regres sion across country was used in
order to analyse the moderation of country on the association between
predictors and the three groups. We investigated a model with gender,
conflict, imbalance of power and support as predictors. In addition, the
interactions between gender and the other predictors and between conflict
and power, given their relevant interplay, were tested for statistical
significance. Since no interaction s between gender and predictors were
found, we deleted them from the final model.
The comparison between the constrained (log likelihood ¼ 7 429.956,
df ¼ 15; scaling correction ¼ 0.981; AIC ¼ 889.912; BIC ¼ 943.777) and the
unconstrained models (log likelihood ¼ 7 421.879, df ¼ 25; scaling
correction ¼ 0.966; AIC ¼ 893.75 8; BIC ¼ 983.532) showed a non-significant
10
MENESINI ET AL.
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
difference in chi-square difference test (Trd ¼ 17.123; df ¼ 10; p ¼ .072),
suggesting the constrained model as the best model and thus the structural
paths are equal across country. Table 3 shows the regression coefficien ts and
the odds ratio for the final model.
Significant positive associations were found for conflict and imbalance of
power in the RPPA profile versus NI. A unit of change in conflict and in
imbalance of power increases the odds of being in the RPPA rather than in
the NI group by about 6 and 7 times, respectively. Significant negative
association was fou nd for support in the RPA profile versus NI. A unit of
change in support decreases the odds of being in the RPA rather than in the
NI group by about 0.64 times. Finally, the comparison between RPPA and
RPA showed a significant positive effect of conflict and imbalance of power
and a significant effect of interaction between conflict and imbalance of
power. With regard to the main effect, RPA are less likely than RPPA to be
associated with conflict and imbalance of power: a unit of change in conflict
and in imbalance of power increases the odds of being in the RPPA rather
than in the RPA group by about 4 and 7 times, respectively. In order to
evaluate the significant effect of the interaction term, we evaluated the odds
ratios (OR) of conflict in two levels of imbalance of power (lower and higher
than 50th percentile). The odds ratio in the first group (N ¼ 245) was
significant and negative (OR ¼ 0.517**). The odds ratio in the second group
(N ¼ 24) was marginally significant and positive (OR ¼ 5.576*). Lower levels
of conflict are more likely to be associ ated with RPA than with RPPA when
the level of imbalance of power is low; on the contrary, with a high level of
imbalance of power, higher levels of conflict are more likely to be associated
TABLE 3
Betas (and standard errors) resulting from multiple-group multinomial logistic
regression (constrained model across Italy and Spain) and odds ratios (OR)
Psychological vs.
Not involved
Psychological and physical vs.
Not involved
Psychological and
physical vs. Psychological
Gender
(M vs. F)
0.571 (0.32);
OR ¼ 1.770
0.143 (0.42);
OR ¼ 1.154
–0.43 (0.42);
OR ¼ 0.651
Conflict 0.288 (0.44);
OR ¼ 1.334
1.86*** (0.59);
OR ¼ 6.424
1.572*** (0.57);
OR ¼ 4.816
Imbalance
of power
–0.063 (0.56);
OR ¼ 0.940
2.008*** (0.74);
OR ¼ 7.448
2.071*** (0.69);
OR ¼ 7.993
Support –0.447*** (0.19);
OR ¼ 0.639
–0.29 (0.26);
OR ¼ 0.748
0.157 (0.25);
OR ¼ 1.170
Conflict 6
Imbalance
of power
0.130 (0.22);
OR ¼ 1.139
–0.49 (0.26);
OR ¼ 0.613
–0.62*** (0.24);
OR ¼ 0.538
Note:*p 5 .05; **p 5 .01; ***p 5 .001.
CROSS-CULTURAL DATING AGGRESSION
11
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
with RPA than RRPA. Overall, when the imbalance of power is low, RPPA
is characterized by higher levels of conflict than RPA: when the imbalance of
power is high, RPPA is characterized by lower levels of conflict than RPA.
DISCUSSION
Literature on dating aggression has shown a pattern of reciprocal
involvement of both partners in perpetrated and received behaviours as
the most frequent type of aggression occurring between young partners.
This result was confirmed in the present study where about 50% of the
adolescents were involved in some reciprocal forms of aggressive acts
toward their partner.
The present study demonstrated that multiple patterns of aggression can
be found in situational couple violence and that some relational factors can
influence this distinction. Although the main feature of situational couple
violence is the reciprocity of the acts and the symmetry of power, distinct
patterns can be found considering the interplay between different levels of
aggression, conflict and imbalance of power. Three theoret ically defined
patterns of involvement in dating aggression—non involvement (NI),
reciprocal involvement in psychological aggression only (RPA), reciprocal
involvement in psychological and physical aggression (RPPA)—differ in
relation to these key variables. Following the family research approach,
results stressed the role of the escalation of conflict as a situational correlate
of aggression between partners (Straus & Gelles, 1990). Partners involved in
RPPA are more likely to have a higher level of conflict and power imbalance
than NI and RPA couples. However, the RPA group did not show any
differences for conflict and imbalance of power as compared to the NI
group. This result can be related to Italian and Spanish cultural habits . It
may be that this behaviour is more ‘‘normative’’ for these two co untries than
for other cultures. A broader and more complete definition of psychological
dating aggression may better explain these results.
Furthermore, RPPA is more likely to ha ve a higher level of conflict than
RPA when the imbalance of power is low. The interaction between conflict
and imbalance of power showed an interesting distinction within RPPA
couples. Som e of them present a high level of conflict and a low level of
imbalance of power, and a minority present high levels of imbalance of
power and low levels of conflict. If the former represents the extreme of the
continuum of a pattern of reciprocal aggression characterized by escalating
levels of conflict and by symmetry in power between partners, the latter may
represent a pattern of aggression where partners fight mainly for power and
control in the relationship. Literature has shown that the aggression is more
likely to be reciprocated when power and control are more balanced
between partners (Olson, 2002). Our resul ts showed that symmetrical
12
MENESINI ET AL.
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
relationships can be characterized by high level of imbalance of power.
However, this condition is not represented by a dominant and submitted
status but rather by an aggressive dynamic where both partners fight for
control during a conflict. The reason for fighting is related to the need to
establish control during a specific conflict or in relation to making a
decision. This finding is in accord with the classification proposed by Ols on
(2002, 2004), where aggressive and violent relationships are distinguished on
the basis of the level of aggression and of the extent to which the partners
fight for control.
The role of suppo rt, defined as communication and expectations for the
future, showed a significant effect in comparing the NI and the RPA groups.
Young people involved in RPA are characterized by lower level of support
than those NI and this is in line with our hypothesis. However, no significant
effects were found in the comparison between the RPPA and the other two
groups. This result was unexpected and may be associated with literature
showing that dating aggression often occurs in long-lasting and more
committed relationships (Capaldi & Crosby, 1997). Reporting a good level
of support despite a high level of aggression seems to respond to the need of
maintaining the relationship.
An additional strength of this study is the comparison of results between
two cities located in two different European countries, Italy and Spain,
where only a little research has been conducted so far. The overall invariance
of latent factors across the two samples allowed us to establish a significant
comparison. Overall , though no significant country effect was found in the
comparison between the involvement in the three groups, other significant
differences were found. First, Spanish adolescents reported higher mean
levels of physical aggression perpetrated and received than Italians. These
results are in line with studies underlining a more consistent involvement in
dating aggression in Spain as compared to other Western countries (Mun
˜
oz-
Rivas et al., 2007b). Second, in relation to the quality of romantic
relationships, the Italian group showed higher levels of power imbalance
and lower levels of support and conflict. Despite the differences at
descriptive level, no significant country-related moderation of the studied
dynamics was found, supporting a generalization of our results for the two
samples. The relevant role that imbalance of power has in either country
could be culturally related: dating partners fight for control during a conflict
or in relation to making a decision.
These results have relevant implications for prevention programmes,
which, for both countries, have to be focused on conflictual dynamics and
on positive strategies to solve them. Though few interventions on dating
aggression are focused on this perspective of reciprocity more efforts are
needed in Europe to tackle the problem by taking into consideration the
perspective of both partners (Graham-Kevan, 2007).
CROSS-CULTURAL DATING AGGRESSION 13
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
Finally, some limitations of the study should be discussed and possibly
overcome in future studies. First, future research has to consider not only
the self-reported point of view on dating aggression but also the perspectives
of both partners, particularly in relation to the reciprocal involvement
profile. Second, a deeper understanding of the imbalance of power should be
pursued considering the structure of symmetry and its direction. Third, the
dimension of support showed a contradictory role, consequently future
studies should progress towards understanding how aggression and support
interact to maintain risk relationships over time. Fourth, the generalizability
of Tuscany’s and Andalusia’s samples needs to be extended to more
representative national samples for each country. Furthermore, studies on
non-Western countries could be relevant in order to understand whether
different society values can affect the dynamics of reciprocal aggressive
young couples. Finally, longitudinal studies can contribute to a more
thoroughly understanding of the meaning and the developm ent of dating
aggression and reciprocal involvement in conflict dyads of adolescents and
young adults in either country.
Manuscript received 14 July 2009
Revised manuscript accepted 29 November 2010
First published online
REFERENCES
Archer, J. (2000). Sex differences in aggression between heterosexual partners: A meta-analytic
review. Psychological Bulletin, 126, 651–680.
Archer, J. (2006). Cross-cultural differences in physical aggression between partners: A social-
structural analysis. Personality and Social Psychology Review, 10, 133–153.
Bollen, K. A. (1989). Structural equations with latent variables (Series in probability and
mathematical statistics). New York, NY: Wiley.
Brown, M. W., & Cudek, R. (1993). Alternative ways of assessing model fit. In K. Bollen & J. S.
Long (Eds.), Testing structural equation models (pp. 136–162). Newbury Park, CA: Sage.
Capaldi, D. M., & Crosby, L. (1997). Observed and reported psychological and physical
aggression in young, at-risk couples. Social Development, 6(2), 184–206.
Capaldi, D. M., & Kim, H. K. (2007). Typological approaches to violence in couples: A critique
and alternative conceptual approach. Clinical Psychological Review, 27(3), 253–265.
Cascardi, M., & Vivian, D. (1995). Context for specific episodes of marital violence: Gender and
severity of violence differences. Journal of Family Violence, 10, 265–293.
Collins, W. C., Raju, N. S., & Edwards, J. E. (2000). Assessing differential functioning in a
satisfaction scale. Journal of Applied Psychology, 85(3), 451–461.
Connolly, J., Nocentini, A., Menesini, E., Pepler, D., Craig, W., & Williams, T. (2010).
Adolescent dating aggression in Canada and Italy: A cross-national comparison.
International Journal of Behavioural Development, 34(2), 98–105.
Crick, N. R. (1995). Relational aggression: The role of intent attributions, provocation type,
and feelings of distress. Development and Psychopathology, 7, 313–322.
Frieze, I. H. (2005). Female violence against intimate partners: An introduction. Psychology of
Women Quarterly, 29, 229–237.
14 MENESINI ET AL.
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011
Furman, W., & Buhrmester, D. (1992). Age and sex differences in perceptions of networks of
personal relationships. Child Development, 63(1), 103–115.
Graham-Kevan, N. (2007). Domestic violence: Research and implications for batterer
programmes in Europe. European Journal on Criminal Policy and Research, 13(3), 213–225.
Gray, H. M., & Foshee, V. (1997). Adolescent dating violence. Journal of Interpersonal
Violence, 12, 126–142.
Henton, J., Cate, R., Koval, J., Lloyd, S., & Christopher, F. S. (1983). Romance and violence in
dating relationships. Journal of Family Issues, 4, 467–482.
Instituto de la Mujer (IM). (2006). Macroencuesta ‘‘Violencia contra las mujeres’’ (Available at:
http://www.mtas.es/mujer/mujeres/cifras/violencia/macroencuesta_violencia.htm)
Italian Institute of Statistics (ISTAT). (2006). La violenza e i maltrattamenti contro le donne
dentro e fuori la famiglia. Rome, Italy: ISTAT.
Johnson, M. P. (1995). Patriarchical terrorism and common couple violence: Two forms of
violence against women. Journal of Marriage and the Family, 57, 283–294.
Kelly, J. B., & Johnson, M. P. (2008). Differentiation among types of intimate partner violence:
Research update and implications for interventions. Family Court Review, 46(3), 476–499.
Menesini, E., & Nocentini, A. (2008). Comportamenti aggressivi nelle prime esperienze
sentimentali in adolescenza. Giornale Italiano di Psicologia, 2, 405–430.
Menesini, E., & Nocentini, A. (2009). Comportamenti aggressivi nel gruppo dei pari e nelle
relazioni sentimentali: quali continuita
`
? Psicologia Clinica dello Sviluppo, 1, 63–80.
Mun
˜
oz-Rivas, M. J., Gran
˜
a, J. L., O’Leary, K. D., & Gonza
´
lez, M. P. (2007a). Aggression in
adolescent dating relationships: Prevalence, justification, and health consequences. Journal
of Adolescent Health, 40(4), 298–304.
Mun
˜
oz-Rivas, M. J., Gran
˜
a, J. L., O’Leary, K. D., & Gonza
´
lez, M. P. (2007b). Physical and
psychological aggression in dating relationships in Spanish university students. Psicothema,
19(1), 102–107.
Muthe
´
n, L. K., & Muthe
´
n, B. O. (2006). Mplus user’s guide (Version 4.1). Los Angeles, CA:
Muthe
´
n & Muthe
´
n.
Nocentini, A., Menesini, E., Pastorelli, C., Connolly, J., Pepler, D., & Graig, W. (in press).
Physical dating aggression in adolescence: cultural and gender invariance. European
Psychologist.
O’Leary, K. D., & Slep, A. M. S. (2003). A dyadic longitudinal model of adolescent dating
aggression.
Journal of Clinical Child and Adolescent Psychology, 32, 314–327.
O’Leary, K. D., & Woodin, E. M. (2009). Psychological and physical aggression in couples:
Causes and interventions. Washington, DC: American Psychological Association.
Olson, L. N. (2002). Exploring ‘‘common couple violence’’ in heterosexual romantic
relationships. Western Journal of Communication, 66, 104–128.
Olson, L. N. (2004). Relational control-motivated aggression: A theoretically based typology of
intimate violence. Journal of Family Communication, 4, 209–233.
Seiffge-Krenke, I., & Connolly, J. (2010). Adolescent romantic relationships across the globe:
Involvement, conflict management and linkages to parents and peer relationships.
International Journal of Behavioral Development, 34(2), 97.
Straus, M. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT)
Scales. Journal of Marriage and the Family, 41, 75–88.
Straus, M. A., & Gelles, R. J. (Eds.). (1990). Physical violence in American families: Risk factors
and adaptations to violence in 8,145 families. New Brunswick, NJ: Transactions.
Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement
invariance literature: Suggestions, practices, and recommendations for organizational
research. Organizational Research Methods, 3, 4–69.
Williams, S. L., & Frieze, I. H. (2005). Patterns of violent relationships, psychological distress,
and marital satisfaction in a national sample of men and women. Sex Roles, 52, 771–784.
CROSS-CULTURAL DATING AGGRESSION
15
Downloaded By: [Menesini, Ersilia] At: 11:34 20 February 2011