- Email: [email protected]
Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents
IJLP-01133; No of Pages 6 International Journal of Law and Psychiatry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Law and Psychiatry
Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents Pedro Pechorro a,⁎, Ricardo Barroso b, Carlos Poiares c, João Pedro Oliveira c, Ohiana Torrealday d a
School of Psychology, University of Minho, Portugal University of Trás-os-Montes e Alto Douro, Portugal Lusófona University of Humanities and Technologies, Portugal d UTMB Correctional Managed Care, USA b c
a r t i c l e
i n f o
Available online xxxx Keywords: Assessment Aggression BPAQ-SF Juvenile delinquency Validation
a b s t r a c t The aim of the present study was to validate the Buss–Perry Aggression Questionnaire-Short Form (BPAQ-SF) among Portuguese juvenile delinquents. With a total sample of 237 male participants, subdivided into an incarcerated forensic sample (n = 192) and a community sample (n = 45), the Portuguese version of the BPAQ-SF demonstrated good psychometric properties in terms of factor structure, internal consistency, convergent validity, discriminant validity, predictive validity and known-groups validity that generally justify its use among Portuguese youth. Statistically significant associations were found with drug use and alcohol abuse. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Aggression is an important risk factor for various forms of antisocial behavior including delinquency and conduct problems, but research has uncovered considerable heterogeneity characterizing the aggression construct (Cima & Raine, 2009). The study of aggression may take different perspectives, focusing on its aims or functions or assessing its various types of manifestations. Some classifications systems of aggression have referred to physical versus verbal aggression (e.g., Buss, 1961), direct versus indirect aggression (e.g., Little, Jones, Henrich, & Hawley, 2003), proactive versus reactive aggression (e.g., Anderson & Bushman, 2002), or instrumental versus impulsive aggression (e.g., Berkowitz, 1993). None of the multiple attempts proposing a taxonomic model valid for aggression have been entirely successful because a comprehensive classification integrating all types of aggression is still to be universally adopted (Parrott & Giancola, 2007). Historically, one of the definitions more widely accepted in the scientific community is that of Buss (1961), who stated that aggression is a response that brings a harmful effect in another organism. Buss and Perry (1992) developed the Aggression Questionnaire (AQ), often referred to as the Buss–Perry Aggression Questionnaire (BPAQ), which has become one of the most popular self-report questionnaires — if not the most popular — for the measurement of aggression since its publication (Morren & Meesters, 2002; Vigil-Colet, Lorenzo-Seva, Codorniu-Raga, & Morales, 2005). In their work, Buss and Perry (1992) revised the original sevenfactor Buss–Durkee Hostility Inventory (BDHI): some items were ⁎ Corresponding author at: School of Psychology, University of Minho, Campus de Gualtar, 4710-057 Braga, Portugal. Tel.: +351 253604267. E-mail address: [email protected] (P. Pechorro).
reworded or omitted and new ones were incorporated, leading to an initial pool of 52 items; 5-point Likert-type scale items replaced the true– false response. They conducted exploratory and confirmatory factor analyses on three separate samples of undergraduate students that led to the empirically derived 29 item four-factor (Physical Aggression, Verbal Aggression, Hostility, and Anger) measure of aggression. The authors reported that their measure had good internal consistency, adequate stability over time, good convergent validity and good discriminant validity. Although the majority of research using the BPAQ has been conducted in English-speaking populations, there is a growing body of research in other cultures. The BPAQ has been translated into several languages, including: Chinese (Maxwell, 2007), Croatian (Mejovsek, Budanovac, & Sucur, 2000), Dutch (Meesters, Muris, Bosma, Schouten, & Beuving, 1996), French (Nahama, Ayoub, Borie, & Petit, 2003), German (Von Collani & Werner, 2005), Italian (Fossati, Maffei, Acquarini, & Di Ceglie, 2003), Japanese (Nakano, 2001), Portuguese (Cunha & Gonçalves, 2012; Simões, 1993), Russian (Ruchkin & Eisemann, 2000), Slovak (Lovas & Trenkova, 1996), Spanish (Andreu, Peña, & Graña, 2002; Santisteban, Alvarado, & Recio, 2007), and Swedish (Prochazka & Agren, 2001). Despite its popularity and the fact it has been widely used in the study of aggression, a number of studies have encountered difficulties when attempting to replicate the original factor structure of the AQ. Some studies reported that the four-factor structure had a poor fit (e.g., Archer, Kilpatrick, & Bramwell, 1995; Harris, 1997; Williams, Boyd, Cascardi, & Poythress, 1996). Other studies found it was possible to obtain a better fit after some items were removed (Harris, 1995; Meesters et al., 1996; Nakano, 2001). Bryant and Smith's (2001) investigation indicated that the four-factor structure of the BPAQ did not explain enough common variance (i.e., about 80%) and lacked the
http://dx.doi.org/10.1016/j.ijlp.2015.08.033 0160-2527/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
2
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx
The forensic sample was recruited from inmates of eight national juvenile detention centers managed by the Portuguese Ministry of Justice. A total sample of 237 male participants (age range = 13–18 years; mean age = 16.61 years; SD = 1.49 years), subdivided into a forensic sample (n = 192; age range = 13–18 years; mean age = 16.62 years; SD = 1.52 years) and a community sample (n = 45; age range = 13– 18 years; mean age = 16.56 years; SD = 1.25 years), agreed to voluntarily participate in the study. The forensic and community participants statistically differed on some moderator variables. The forensic sample had more ethnic minorities participants (χ2 = 36.262, p ≤ .001), fewer participants from urban areas (χ2 = 61.739, p ≤ .001), fewer years of education (t = 17.785, p ≤ .001), lower parental socio-economic status (χ2 = 21.464, p ≤ .001), more divorced or deceased parents (χ2 = 63.763, p ≤ .001), and more siblings/half-siblings (t = −3.655, p ≤ .001). The forensic and community participant samples did not differ in a statistically significant manner either by age or nationality.
on a 5-point Likert scale (from Extremely uncharacteristic of me = 1 to Extremely characteristic of me = 5) that provides a global measure of aggression and four subscales: Physical Aggression (PA, 9 items), Verbal Aggression (VA, 5 items), Anger (A, 7 items), and Hostility (H, 8 items). The Buss–Perry Aggression Questionnaire-Short Form (BPAQ-SF; Bryant & Smith, 2001) is a “refined” version of the AQ consisting of 12 Likerttype items rated on a 6-point ordinal scale. The BPAQ-SF is also organized into four scales: Physical Aggression (PA, 3 items), Verbal Aggression (VA, items), Anger (A, 3 items), and Hostility (H, 3 items). Bryant and Smith (2001) decided to change the original 5-point scale to a 6-point scale to eliminate the scale's midpoint and force respondents to decide whether each statement was characteristic of them. Our Portuguese version consisted of Likert-type items rated on a 5-point ordinal scale (Never = 0 to Always = 4), restoring the scale's mid-point which was eliminated in Bryant and Smith's version, because five response alternatives with a neutral point seem to be more adequate in the context of cross-cultural comparisons and the reliability of a personality measurement instrument does not vary significantly when one response category is reduced (Gallardo-Pujol, Kramp, García-Forero, Pérez-Ramírez, & Andrés-Pueyo, 2006). The Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1989) is a brief self-report measure that evaluates self-esteem in adolescents and adults. The RSES can be scored by simply adding the ten items on a 4-point ordinal scale (Strongly Disagree = 0, Disagree = 1, Agree = 2, Strongly Agree = 3) after reverse scoring designated items (namely, items 2, 5, 6, 8 and 9). Higher scores indicate higher levels of selfesteem. A Portuguese version of the RSES was used (Pechorro, 2011). Pechorro, Marôco, Poiares, and Vieira (2011) found psychometric properties that justify the use of the RSES with Portuguese adolescent community and forensic populations, namely, in terms of internal consistency (Cronbach's α = .79), unidimensional factor structure (35.55% of variance), temporal stability (rs = .86; p ≤ .01), known-groups validity (Λ Wilks = .961; χ2 = 29.806; p ≤ .001), discriminant validity (r = .10; ns), corrected item–total correlation (range = .27–.62.) and average inter-item correlation (.27). Internal consistency for the present study, estimated by Cronbach's alpha, was .77. The Reactive–Proactive Aggression Questionnaire (RPQ; Raine et al., 2006) is a self-report measure that distinguishes between reactive and proactive aggression. The scale consists of 23 items rated on a 3-point ordinal scale (Never = 0, Sometimes = 1, Often = 2). A total of 11 items assess reactive aggression (e.g., “Reacted angrily when provoked by others”) and 12 items assess proactive aggression (e.g., “Hurt others to win a game”). Summed scores provide a measures of reactive or proactive aggression, as well as total aggression. Higher scores indicate higher levels of aggression. The RPQ is appropriate for use with youth in late adolescence and young adults. Internal consistency for adolescents has previously been reported as .86 for proactive aggression, .84 for reactive aggression, and .90 for total aggression (Raine et al., 2006). The Portuguese form of the RPQ was used (Pechorro, Ray, Raine, Maroco, & Gonçalves, in press) in this study. The internal consistency for the present study, estimated by Cronbach's alpha, was as follows: RPQ Reactive = .81; RPQ Proactive = .83; RPQ Total = .86. In addition, a questionnaire was constructed to describe the sociodemographic and criminal characteristics of the participants and to analyze the possible moderating effect of these variables. This questionnaire included variables such as participants' age, nationality, ethnic group, origin (rural vs. urban), level of schooling completed, socioeconomic status, parental marital status, number of siblings/half-siblings, previous use of physical violence, drug use and alcohol abuse. Socioeconomic status was measured by considering both parental level of education and profession, appropriate to the Portuguese reality (Simões, 1994).
2.2. Measures
2.3. Procedures
The Buss–Perry Aggression Questionnaire (BPAQ; Buss & Perry, 1992) is a four-factor model questionnaire consisting of 29 items scored
Appropriate procedures were followed during the translation and retroversion (Hambleton, Merenda, & Spielberger, 2005) of the
specificity required of this type of instrument to serve as a measurement model. Using confirmatory factor analysis (CFA) they found that Buss and Perry's original four-factor model achieved only mediocre fit in their three samples (Goodness of Fit Index [GFI] = .76–.81; χ2/df = 2.4–2.8). This state of affairs inspired Bryant and Smith (2001) to develop a reduced version of the BPAQ using five independent data sets. These authors removed all the items that loaded less than .40 or had crossed loadings (i.e., more than .40 in two or more components). This produced a shortened questionnaire consisting of three items for each factor. This modification resulted in a good fit to the four-factor model in several samples and explained an acceptable proportion of variance in both absolute and relative terms (GFI = .94). The reliabilities of the refined factors were corrected for differences in the number of constituent items (adjusted α's = .88–.92) using the Spearman–Brown prophecy formula, and were considered good. The data also provided strong support for the convergent and discriminant validity of the refined Physical Aggression, Anger, and Hostility factors, while the Verbal Aggression factor, showed poor discriminant validity. The authors concluded that their 12-item refined version (often referred to as BPAQ-SF) demonstrated superior psychometric properties in its overall goodness-of-fit to the data, yet still retained the conceptual framework originally proposed. The BPAQ and its short forms (Bryant & Smith, 2001; Buss & Perry, 1992; Buss & Warren, 2000) have proven their value in studying aggression profiles and predicting violent behavior in adults (e.g., Bushman & Wells, 1998; Diamond & Magaletta, 2006; Diamond, Wang, & Buffington-Vollum, 2005), while the same cannot be said about adolescents and preadolescents. The trajectory of developmental aggression (e.g., low, persistent or desisting aggression) among youth (e.g. Collishaw, Maughan, Goodman, & Pickles, 2004; Martino, Ellickson, Klein, McCaffrey, & Edelen, 2008), from late childhood through adolescence, needs valid measurement. The adaptation of the BPAQ-SF to youth within this age range would be very useful in terms of obtaining a short and valid measurement of the aggression trait. The main aim of the present study is to validate a Portuguese version of the Buss–Perry Aggression Questionnaire-Short Form (BPAQ-SF; Bryant & Smith, 2001) so as to further the investigation of aggression among juvenile delinquents and community youths in Portugal and Portuguese speaking countries. 2. Method 2.1. Participants
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx
instrument. The initial translation from English into Portuguese was completed by the first and second authors of this article, who made sure that young people would be able to properly understand the meaning of the items. The questionnaire was then independently translated back into English by a native English speaker with considerable professional experience in translating psychology-related scientific texts. No significant differences were found between the back-translation and the original version, demonstrating that the translated items had the same or very similar meanings as the original English items. The original items and the Portuguese translation are available in Appendix A and Appendix B, respectively. Authorization to assess youths was obtained from the General Directorate of Reintegration and Prison Services, Ministry of Justice (DireçãoGeral de Reinserção e Serviços Prisionais, Ministério da Justiça). The detainees from the existing male Juvenile Detention Centers were informed about the nature of the study and asked to participate. The measures were administered by means of individual face-to-face interviews in an appropriate setting. The majority of the participants were convicted of serious and violent crimes (e.g., homicide, robbery, assault, rape, child sexual abuse), and were detained by the court's decision. Authorization was also obtained from the Directorate General of Education, Ministry of Education (Direção-Geral de Educação, Ministério da Educação). The community sample was recruited from schools in the Lisbon region (Portugal) that agreed to participate. Written consent was required from each participating student's parent or guardian. Participants who were unwilling or unable to collaborate were excluded. The data was analyzed using SPSS v22 (IBM SPSS, 2013) and EQS 6.2 (Bentler & Wu, 2008). Following data entry 15% of the questionnaires were randomly selected to evaluate accuracy of data input, which was judged to be good. The factor structure of the Portuguese language version of the BPAQ was assessed with Confirmatory Factor Analysis (CFA) performed in EQS 6.2 (Bentler & Wu, 2008; Byrne, 2006), with robust estimation methods appropriate for ordinal variables and evaluation of model fit. The goodness of fit indices calculated included chi-square/ degrees of freedom, Satorra–Bentler chi-square/degrees of freedom, comparative fit index (CFI) and root mean square error of approximation (RMSEA). A chi-square/degrees of freedom value b 5 is considered adequate, ≤2 is considered good and 1 = very good (Marôco, 2014; West, Taylor, & Wu, 2012). A CFI ≥ .90 and RMSEA ≤ .10 indicate adequate fit, whereas a CFI ≥ .95 and RMSEA ≤ .06 indicate good model fit (Byrne, 2006). Items with standardized loading above .45 were retained. The chi-square test was used to compare nominal variables. Pearson's r was used to analyze correlations between scale variables. Point-bisserial correlations were used to analyze the association between nominal dichotomous variables and scale variables. Discriminant analysis was used to assess known-groups validity. Results were considered significant if p ≤ .05 and marginally significant if p ≤ .1 (Aron, Coups, & Aron, 2013).
3. Results The aim of our study was to validate a short version of the Buss– Perry Aggression Questionnaire for Portuguese youths. Our first step in attempting to validate the Portuguese version of the BPAQ-SF was to try to confirm the four-factor structure obtained in previous studies carried out using the EQS 6.2 software (Bentler & Wu, 2008). No support was found for the original BPAQ by means of CFA using the ML method due to poor fits and/or low item loadings. The BPAQ-SF four-factor structure showed a better fit, but using EQS's ML Robust method for skewed data we were able to obtain the best fit for the BPAQ-SF (see Table 1). In Fig. 1 are the loadings for the four-factor structure of the BPAQ-SF estimated with the ML Robust method. Set in Table 2 are the correlation matrix values.
3
Table 1 Goodness of fit indexes for the different models.
BPAQ 4-factor BPAQ-SF 4-factor BPAQ-SF 4-factor Robust
χ2/df or S-Bχ2/df
CFI
RMSEA (90% CI)
2.86 2.94 1.43
.80 .97 .99
.07 (.06, .07) .07 (.06, .08) .03 (.01, .05)
Note. BPAQ = Aggression Questionnaire; BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; χ2 = chi-square; S-Bχ2 = Satorra–Bentler chi-square; df = degrees of freedom; CFI = comparative fit index; RMSEA (90% CI) = root mean square error of approximation (90% confidence interval); ML = maximum likelihood.
The following step was the estimation of Cronbach's alpha, mean inter-item correlation and corrected item–total correlation range (see Table 3). The convergent validity of the BPAQ-SF with the RPQ and its dimensions revealed a statistically significant positive correlation, while the discriminant validity with the RSES revealed a statistically significant negative correlation (see Table 4). The predictive validity of the BPAQ-SF score with previous physical violence (coded No = 0, Yes = 1) revealed the existence of a statistically significant positive point-bisserial correlation. Similarly, the associations of the BPAQ-SF with criterion-related variables such as drug use (coded No = 0, Yes = 1) and alcohol abuse (coded No = 0, Yes = 1) also revealed the existence of statistical significant positive pointbisserial correlations (see Table 5). The known-groups validity of the BPAQ-SF was calculated using Wilks' lambda; statistically significant differences were found between the forensic sample and the community sample (Wilks' λ = .96, χ2 = 5.147 (1), p ≤ .05). 4. Discussion The present study had as its aim the analysis of the psychometric properties of the BPAQ-SF, a shortened and refined version of the measure that is considered the gold standard in the assessment of aggression, among Portuguese youth. In terms of factor structure, no support was found for the original BPAQ by means of CFA using the ML method due to poor fits. This was somewhat expected because some of the items had non-normal distributions. The BPAQ-SF showed an adequate fit using the same ML method. However, by using the ML Robust method developed for skewed data, we were able to obtain a very good fit confirming the four-factor structure of the BPAQ-SF. Evidence obtained in our study shows that Bryant and Smith's (2001) refined four-factor model fits our data very well. Some psychometric problems were found when analyzing our Portuguese adaptation of the BPAQ-SF. Analysis of the internal consistency revealed somewhat low values for the Verbal Aggression (VA) and the Anger (A) factors that may put into question the reliability of measurement for these two factors (Cortina, 1993; Kaplan & Saccuzzo, 2009). These values were lower than the reliability estimates obtained by Bryant and Smith (2001), which ranged from .79 to .80 for physical aggression, .73 to .83 for verbal aggression, .71 to .76 for anger, and .70 to .75 for hostility. Because Cronbach's alpha is sensitive to the number of items included in a given scale, these authors also calculated adjusted alpha coefficients of .88 to .92 for the four subscales. Alpha values for an overall aggressiveness score were not provided. There were also some positive findings. Regarding the mean interitem correlations (which is a straightforward measure of internal consistency), values were within the recommended range of .15–.50 (Clark & Watson, 1995; Domino & Domino, 2006). In terms of the corrected item–total correlation range, the BPAQ-SF total score and its four factors all reached the minimum recommended value of .20 (Kaplan & Saccuzzo, 2009; Nunnally & Bernstein, 1994). The convergent validity of the BPAQ-SF with the RPQ and its two factors revealed a moderate–high to strong statistically significant correlation, while the discriminant validity with the RSES revealed a
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
4
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx
Fig. 1. BPAQ-SF four-factor ML Robust structure with standardized item loadings.
moderate–low negative correlation, demonstrating the expected construct overlap of the former but not of the latter (American Psychological
Table 2 Correlation matrix.
1 PA 2 VA 3A 4H 5 BPAQ-SF
1 BPAQ-SF PA
2 BPAQ-SF VA
1 .88⁎ .79⁎ .50⁎ .79⁎
1 .94⁎ .45⁎ .78⁎
3 BPAQ-SF A
4 BPAQ-SF H
5 BPAQ-SF total
Association, 1999; Kaplan & Saccuzzo, 2009). This is in line with those found in previous studies (Bryant & Smith, 2001; Williams et al., 1996). As expected (American Psychological Association, 1999), and in keeping with previous studies (e.g. Diamond & Magaletta, 2006), a Table 3 Cronbach's alpha, mean inter-item correlation, and corrected item–total correlation range. BPAQ-SF PA BPAQ-SF VA BPAQ-SF A BPAQ-SF H BPAQ-SF total
1 .64⁎ .81⁎
1 .70⁎
Cronbach's α .71 MIIC .45 CITCR .49–.55 1
Note. BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; BPAQ-SF PA = Physical Aggression dimension; BPAQ-SF VA = Verbal Aggression dimension; BPAQ-SF A = Anger dimension; BPAQ-SF H = Hostility dimension ⁎ Significant at the .05 level.
.63 .36 .37–.48
.62 .36 .37–.47
.75 .49 .50–.63
.84 .30 .40–.59
Note. BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; BPAQ-SF PA = Physical Aggression dimension; BPAQ-SF VA = Verbal Aggression dimension; BPAQ-SF A = Anger dimension; BPAQ-SF H = Hostility dimension; Cronbach's α = Cronbach's alpha; MIIC = mean inter-item correlation; CITCR = corrected item–total correlation range.
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx Table 4 Convergent validity with RPQ Reactive, RPQ Proactive and RPQ Total, and discriminant validity with RSES.
BPAQ-SF
RPQ Reactive
RPQ Proactive
RPQ Total
RSES
r = .61⁎⁎⁎
r = .51⁎⁎⁎
r = .65⁎⁎⁎
r = −.38⁎⁎⁎
Note. BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; RPQ = Reactive–Proactive Aggression Questionnaire; RPQ Proactive = Proactive dimension; RPQ Reactive = Reactive dimension; RSES = Rosenberg Self-Esteem Scale. ⁎⁎⁎ Significant at the .001 level.
Table 5 Correlations of BPAQ-SF with previous use of physical violence, drug use, and alcohol abuse.
BPAQ-SF
PPV
DU
AB
rpb = .25⁎⁎
rpb = .20⁎
rpb = .17⁎
Note. BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; PPV = previous physical violence; DU = drug use; AB = alcohol abuse; rpb = point-bisserial correlation. ⁎⁎ Significant at the .01 level. ⁎ Significant at the .05 level.
5
Appendix A Original items of the BPAQ-SF. Items 1 (2). Given enough provocation, I may hit another person. 2 (6). There are people who pushed me so far that we came to blows. 3 (8). I have threatened people I know. 4 (11). I often find myself disagreeing with people. 5 (13). I can’t help getting into arguments when people disagree with me. 6 (14). My friends say that I’m somewhat argumentative. 7 (15). I flare up quickly but get over it quickly. 8 (20). Sometimes I fly off the handle for no good reason. 9 (21). I have trouble controlling my temper. 10 (23). At times I feel I have gotten a raw deal out of life. 11 (24). Other people always seem to get the breaks. 12 (25). I wonder why sometimes I feel so bitter about things.
Appendix B Portuguese translation of the BPAQ-SF. Items
moderate positive correlation was found between the predictive validity of the BPAQ-SF score with a history of previous physical violence. The correlations of the BPAQ-SF with criterion-related variables revealed the existence of moderate–low associations with drug use and alcohol abuse. Associations between aggression scores and drug use and alcohol abuse have been consistently reported in the literature (e.g., McMurran et al., 2006; Taylor & Hulsizer, 1998), although the ones we obtained were somewhat lower but still statistically significant. Also consistent with prior studies (e.g. Smith & Waterman, 2006), the known-groups validity revealed that the BPAQ-SF could significantly discriminate between the forensic group and the community group, conceptualized as structurally different and mutually exclusive (Kaplan & Saccuzzo, 2009). Overall, our findings provide some additional support for Buss and Perry's (1992) four-factor model of aggression among adolescents and its potential generalization across different languages and cultures. It was possible to demonstrate some appropriate psychometric properties that justify the future use of the BPAQ-SF with the Portuguese youth population. To our knowledge this is the first study attempting the validation of the BPAQ-SF among juvenile delinquents in Portugal. Additional validation procedures are recommended and should be done in the future (e.g., test–retest reliability, cross-validation using other samples). Given the enormous costs that violent offenders create in terms of their crimes on victims and society as a whole, we hope our study may guide the use of the BPAQ-SF with Portuguese youth and contribute to the betterment of treatment programs targeting serious and violent juvenile offenders in Portugal.
Acknowledgments We wish to thank the staffs of the following Portuguese juvenile detention centers for their collaboration in this study: Belavista, Mondego, Navarro de Paiva, Olivais, Padre António Oliveira, Santo António, Santa Clara, and Prisão-Escola de Leiria. This study was partly supported by the Portuguese Foundation for Science and Technology (FCT; Grant SFRH/BPD/86666/2012) with co-financing of the European Social Fund (POPH/FSE), the Portuguese Ministry of Education and Science (MEC) through national funds, and co-financed by FEDER under the PT2020 Partnership Agreement (UID/PSI/01662/2013).
1 (2). Se me provocarem bastante, posso bater noutra pessoa. 2 (6). Houve pessoas que me pressionaram tanto que andámos à pancada (luta). 3 (8). Já ameacei pessoas que conheço. 4 (11). Entro em desacordo com as pessoas. 5 (13). Não consigo deixar de discutir quando as pessoas discordam de mim. 6 (14). Os meus amigos dizem que gosto de discutir. 7 (15). Exalto-me facilmente, mas passa-me rapidamente. 8 (20). Perco o controlo sem razão em especial. 9 (21). Tenho dificuldade em controlar o meu feitio. 10 (23). Sinto que a vida não me dá boas oportunidades. 11 (24). As outras pessoas parecem ter sempre as melhores oportunidades. 12 (25). Pergunto-me por que motivo às vezes me sinto tão amargo com as coisas.
References American Psychological Association (1999). Standards for educational and psychological tests. Washington, DC: Author. Anderson, C., & Bushman, B. (2002). Human aggression. Annual Review of Psychology, 53, 27–51. Andreu, J., Peña, M., & Graña, J. (2002). Adaptación psicométrica de la versión española del cuestionario de agresión [Psychometric adaptation of the Spanish version of the Aggression Questionnaire]. Psicothema, 14, 476–482. Archer, J., Kilpatrick, G., & Bramwell, R. (1995). Comparison of two aggression inventories. Aggressive Behavior, 21, 371–380. Aron, A., Coups, E., & Aron, E. (2013). Statistics for psychology (6th ed.). New York, NY: Pearson. Bentler, P., & Wu, E. (2008). EQS for Windows user's guide. Encino, CA: Multivariate Software, Inc. Berkowitz, L. (1993). Aggression: Its causes, consequences and control. Philadelphia: Temple University Press. Bryant, F., & Smith, B. (2001). Refining the architecture of aggression: A measurement model for the Buss–Perry Aggression Questionnaire. Journal of Research in Personality, 35, 138–167. Bushman, B., & Wells, G. (1998). Trait aggressiveness and hockey penalties: Predicting hot tempers on the ice. Journal of Personality and Social Psychology, 83, 969–974. Buss, A. (1961). The psychology of aggression. New York: John Wiley & Sons, Inc. Buss, A., & Perry, M. (1992). The Aggression Questionnaire. Journal of Personality and Social Psychology, 63(3), 452–459. Buss, A., & Warren, W. (2000). The Aggression Questionnaire manual. Los Angeles: Western Psychological Services. Byrne, B. (2006). Structural equation modeling with EQS: Basic concepts, applications, and programming. Mahwah, NJ: Lawrence Erlbaum Associates. Cima, M., & Raine, A. (2009). Distinct characteristics of psychopathy relate to different subtypes of aggression. Personality and Individual Differences, 47, 835–840. Clark, L., & Watson, D. (1995). Constructing validity: Basic issues in objective scale development. Psychological Assessment, 7(3), 309–319. Collishaw, S., Maughan, B., Goodman, R., & Pickles, A. (2004). Time trends in adolescent mental health. Journal of Child Psychology & Psychiatry, 45, 1350–1362. Cortina, J. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78(1), 98–104. Cunha, O., & Gonçalves, R. (2012). Análise confirmatória fatorial de uma versão portuguesa do Questionário de Agressividade de Buss–Perry [Confirmatory factor analysis of a Portuguese version of the Buss–Perry Aggression Questionnaire]. Laboratório de Psicologia, 10(1), 3–17.
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
6
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx
Diamond, P., & Magaletta, P. (2006). The Short-Form Buss–Perry Aggression Questionnaire (BPAQ-SF): A validation study with federal offenders. Assessment, 13(3), 227–240. Diamond, P., Wang, E., & Buffington-Vollum, J. (2005). Factor structure of the Buss–Perry Aggression Questionnaire (BPAQ) with mentally ill male prisoners. Criminal Justice and Behavior, 32(5), 546–564. Domino, G., & Domino, M. (2006). Psychological testing: An introduction (2nd ed.). New York: Cambridge University Press. Fossati, A., Maffei, C., Acquarini, E., & Di Ceglie, A. (2003). Multigroup confirmatory component and factor analyses of the Italian version of the Aggression Questionnaire. European Journal of Psychological Assessment, 19, 54–65. Gallardo-Pujol, D., Kramp, U., García-Forero, C., Pérez-Ramírez, M., & Andrés-Pueyo, A. (2006). Assessing aggressiveness quickly and efficiently: The Spanish adaptation of Aggression Questionnaire-Refined version. European Psychiatry, 21(7), 487–494. Hambleton, R., Merenda, P., & Spielberger, C. (2005). Adapting educational and psychological tests for cross-cultural assessment. Mahwah, NJ: Lawrence Erlbaum Associates. Harris, J. (1995). Confirmatory factor analysis of the aggression questionnaire. Behavioral Research and Therapy, 8, 991–993. Harris, J. A. (1997). A further evaluation of the Aggression Questionnaire: Issues of validity and reliability. Behavior Research and Therapy, 35, 1047–1053. IBM SPSS (2013). IBM SPSS statistics base 22. Chicago, IL: SPSS. Kaplan, R., & Saccuzzo, D. (2009). Psychological testing: Principles, applications, and issues. Belmont, CA: Wadsworth, Cengage Learning. Little, T., Jones, S., Henrich, C., & Hawley, P. (2003). Disentangling the ‘whys’ from the ‘whats’ of aggressive behaviour. International Journal of Behavioral Development, 27, 122–133. Lovas, L., & Trenkova, S. (1996). Aggression perception of an incident. Studia Psychologia, 38, 265–270. Marôco, J. (2014). Análise de Equações Estruturais: Fundamentos teóricos, software & aplicações [Structural equations analysis: Theoretical foundations, software and applications]. Pero Pinheiro: ReportNumber, Ltd. Martino, S., Ellickson, P., Klein, D., McCaffrey, D., & Edelen, M. (2008). Multiple trajectories of physical aggression among adolescent boys and girls. Aggressive Behavior, 34, 61–75. Maxwell, J. (2007). Development and preliminary validation of a Chinese version of the Buss–Perry Aggression Questionnaire in a population of Hong Kong Chinese. Journal of Personality Assessment, 88(3), 284–294. McMurran, M., Egan, V., Cusens, B., Van den Bree, M., Austin, E., & Charlesworth, P. (2006). The alcohol-related Aggression Questionnaire. Addiction Research and Theory, 14(3), 323–343. Meesters, C., Muris, P., Bosma, H., Schouten, E., & Beuving, S. (1996). Psychometric evaluation of the Dutch version of the Aggression Questionnaire. Behavioral Research Therapy, 34, 839–843. Mejovsek, M., Budanovac, A., & Sucur, Z. (2000). The relationship between inmates' aggression and their socioeconomic and family characteristics. Hrvatska Revija Za Rehabilitacijska Istrazivanja, 36, 75–86. Morren, M., & Meesters, C. (2002). Validation of the Dutch version of the Aggression Questionnaire in adolescent male offenders. Aggressive Behavior, 28, 87–96. Nahama, V., Ayoub, M. P., Borie, R., & Petit, F. (2003). Problemes d’adaptation sociale chez des jeunes de 9 a 12 ans presentant des troubles des apprentissages [Social adjustment problems of youngsters with learning difficulties]. Approche Neuropsychologique des Apprentissages chez l’Enfant, 15, 9–13. Nakano, K. (2001). Psychometric evaluation on the Japanese adaptation of the Aggression Questionnaire. Behavioral Research Therapy, 39, 853–858. Nunnally, J., & Bernstein, I. (1994). Psychometric theory. New York, NY: McGraw-Hill.
Parrott, D., & Giancola, P. (2007). Addressing “The criterion problem” in the assessment of aggressive behavior: Development of a new taxonomic system. Aggression and Violent Behavior, 12, 280–299. Pechorro, P. (2011). Juvenile delinquency: Study of some psychological and relational variables with an emphasis on psychopathic traits (Unpublished doctoral dissertation). Faculdade de Medicina da Universidade de Lisboa, Lisbon, Portugal. Pechorro, P., Marôco, J., Poiares, C., & Vieira, R. (2011). Validação da Escala de Auto-Estima de Rosenberg com adolescentes portugueses em contexto forense e escolar [Validation of the Rosenberg Self-Esteem Scale with Portuguese adolescents in forensic and school contexts]. Arquivos de Medicina, 25(5/6), 174–179. Pechorro, P., Ray, J., Raine, A., Maroco, J., & Gonçalves, R. (2015). The Reactive–Proactive Aggression Questionnaire: Validation among a Portuguese sample of incarcerated juvenile delinquents. Journal of Interpersonal Violence, http://dx.doi.org/10.1177/ 0886260515590784 (in press). Prochazka, H., & Agren, H. (2001). Aggression in the Swedish population measured with a new self-rating inventory: The Aggression Questionnaire-revised Swedish version (AQ-RSV). Nordic Journal of Psychiatry, 55, 17–24. Raine, A., Dodge, K., Loeber, R., Gatzke-Kopp, L., Lynam, D. R., Reynolds, C., et al. (2006). The reactive-proactive aggression questionnaire: Differential correlates of reactive and proactive aggression in adolescent boys. Aggressive Behavior, 32, 159–172. Rosenberg, M. (1989). Society and the adolescent self-image (Rev. ed.). Middletown, CT: Wesleyan University Press. Ruchkin, V., & Eisemann, M. (2000). Aggression and psychological problems in juvenile male delinquents versus controls in Russia: Alternate ways of ‘letting off steam’? Aggression and Violent Behavior, 5, 217–225. Santisteban, C., Alvarado, J., & Recio, P. (2007). Evaluation of a Spanish version of the Buss and Perry aggression questionnaire: Some personal and situational factors related to the aggression scores of young subjects. Personality and Individual Differences, 42, 1453–1465. Simões, A. (1993). São os homens mais agressivos que as mulheres? [Are men more aggressive than women?]. Revista Portuguesa de Pedagogia, XXVII(3), 387–404. Simões, M. (1994). Investigação no âmbito da aferição nacional ao Teste das Matrizes Progressivas Coloridas de Raven [Researches regarding the national validation of Raven's Coloured Progressive Matrices Test] (Unpublished doctoral dissertation). Faculdade de Psicologia e Ciências da Educação da Universidade de Coimbra, Coimbra, Portugal. Smith, P., & Waterman, M. (2006). Self-reported aggression and impulsivity in forensic and non-forensic populations. Journal of Family Violence, 21, 425–437. Taylor, S., & Hulsizer, M. (1998). Psychoactive drugs and human aggression. In R. Geen, & E. Donnerstein (Eds.), Human aggression: Theories, research, and implications for social policy (pp. 139–165). New York: Academic Press. Vigil-Colet, A., Lorenzo-Seva, U., Codorniu-Raga, M., & Morales, F. (2005). Factor structure of the Buss–Perry Aggression Questionnaire in different samples and languages. Aggressive Behavior, 31, 601–608. Von Collani, G., & Werner, R. (2005). Self-related and motivational constructs as determinants of aggression. An analysis and validation of a German version of the Buss–Perry aggression questionnaire. Personality and Individual Differences, 38, 1631–1643. West, S., Taylor, A., & Wu, W. (2012). Model fit and model selection in structural equation modeling. In R. Hoyle (Ed.), Handbook of structural equation modeling (pp. 209–231). New York, NY: Guilford Press. Williams, T., Boyd, J., Cascardi, M., & Poythress, N. (1996). Factor structure and convergent validity of the aggression questionnaire in an offender population. Psychological Assessment, 4, 398–403.
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
Contents lists available at ScienceDirect
International Journal of Law and Psychiatry
Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents Pedro Pechorro a,⁎, Ricardo Barroso b, Carlos Poiares c, João Pedro Oliveira c, Ohiana Torrealday d a
School of Psychology, University of Minho, Portugal University of Trás-os-Montes e Alto Douro, Portugal Lusófona University of Humanities and Technologies, Portugal d UTMB Correctional Managed Care, USA b c
a r t i c l e
i n f o
Available online xxxx Keywords: Assessment Aggression BPAQ-SF Juvenile delinquency Validation
a b s t r a c t The aim of the present study was to validate the Buss–Perry Aggression Questionnaire-Short Form (BPAQ-SF) among Portuguese juvenile delinquents. With a total sample of 237 male participants, subdivided into an incarcerated forensic sample (n = 192) and a community sample (n = 45), the Portuguese version of the BPAQ-SF demonstrated good psychometric properties in terms of factor structure, internal consistency, convergent validity, discriminant validity, predictive validity and known-groups validity that generally justify its use among Portuguese youth. Statistically significant associations were found with drug use and alcohol abuse. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Aggression is an important risk factor for various forms of antisocial behavior including delinquency and conduct problems, but research has uncovered considerable heterogeneity characterizing the aggression construct (Cima & Raine, 2009). The study of aggression may take different perspectives, focusing on its aims or functions or assessing its various types of manifestations. Some classifications systems of aggression have referred to physical versus verbal aggression (e.g., Buss, 1961), direct versus indirect aggression (e.g., Little, Jones, Henrich, & Hawley, 2003), proactive versus reactive aggression (e.g., Anderson & Bushman, 2002), or instrumental versus impulsive aggression (e.g., Berkowitz, 1993). None of the multiple attempts proposing a taxonomic model valid for aggression have been entirely successful because a comprehensive classification integrating all types of aggression is still to be universally adopted (Parrott & Giancola, 2007). Historically, one of the definitions more widely accepted in the scientific community is that of Buss (1961), who stated that aggression is a response that brings a harmful effect in another organism. Buss and Perry (1992) developed the Aggression Questionnaire (AQ), often referred to as the Buss–Perry Aggression Questionnaire (BPAQ), which has become one of the most popular self-report questionnaires — if not the most popular — for the measurement of aggression since its publication (Morren & Meesters, 2002; Vigil-Colet, Lorenzo-Seva, Codorniu-Raga, & Morales, 2005). In their work, Buss and Perry (1992) revised the original sevenfactor Buss–Durkee Hostility Inventory (BDHI): some items were ⁎ Corresponding author at: School of Psychology, University of Minho, Campus de Gualtar, 4710-057 Braga, Portugal. Tel.: +351 253604267. E-mail address: [email protected] (P. Pechorro).
reworded or omitted and new ones were incorporated, leading to an initial pool of 52 items; 5-point Likert-type scale items replaced the true– false response. They conducted exploratory and confirmatory factor analyses on three separate samples of undergraduate students that led to the empirically derived 29 item four-factor (Physical Aggression, Verbal Aggression, Hostility, and Anger) measure of aggression. The authors reported that their measure had good internal consistency, adequate stability over time, good convergent validity and good discriminant validity. Although the majority of research using the BPAQ has been conducted in English-speaking populations, there is a growing body of research in other cultures. The BPAQ has been translated into several languages, including: Chinese (Maxwell, 2007), Croatian (Mejovsek, Budanovac, & Sucur, 2000), Dutch (Meesters, Muris, Bosma, Schouten, & Beuving, 1996), French (Nahama, Ayoub, Borie, & Petit, 2003), German (Von Collani & Werner, 2005), Italian (Fossati, Maffei, Acquarini, & Di Ceglie, 2003), Japanese (Nakano, 2001), Portuguese (Cunha & Gonçalves, 2012; Simões, 1993), Russian (Ruchkin & Eisemann, 2000), Slovak (Lovas & Trenkova, 1996), Spanish (Andreu, Peña, & Graña, 2002; Santisteban, Alvarado, & Recio, 2007), and Swedish (Prochazka & Agren, 2001). Despite its popularity and the fact it has been widely used in the study of aggression, a number of studies have encountered difficulties when attempting to replicate the original factor structure of the AQ. Some studies reported that the four-factor structure had a poor fit (e.g., Archer, Kilpatrick, & Bramwell, 1995; Harris, 1997; Williams, Boyd, Cascardi, & Poythress, 1996). Other studies found it was possible to obtain a better fit after some items were removed (Harris, 1995; Meesters et al., 1996; Nakano, 2001). Bryant and Smith's (2001) investigation indicated that the four-factor structure of the BPAQ did not explain enough common variance (i.e., about 80%) and lacked the
http://dx.doi.org/10.1016/j.ijlp.2015.08.033 0160-2527/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
2
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx
The forensic sample was recruited from inmates of eight national juvenile detention centers managed by the Portuguese Ministry of Justice. A total sample of 237 male participants (age range = 13–18 years; mean age = 16.61 years; SD = 1.49 years), subdivided into a forensic sample (n = 192; age range = 13–18 years; mean age = 16.62 years; SD = 1.52 years) and a community sample (n = 45; age range = 13– 18 years; mean age = 16.56 years; SD = 1.25 years), agreed to voluntarily participate in the study. The forensic and community participants statistically differed on some moderator variables. The forensic sample had more ethnic minorities participants (χ2 = 36.262, p ≤ .001), fewer participants from urban areas (χ2 = 61.739, p ≤ .001), fewer years of education (t = 17.785, p ≤ .001), lower parental socio-economic status (χ2 = 21.464, p ≤ .001), more divorced or deceased parents (χ2 = 63.763, p ≤ .001), and more siblings/half-siblings (t = −3.655, p ≤ .001). The forensic and community participant samples did not differ in a statistically significant manner either by age or nationality.
on a 5-point Likert scale (from Extremely uncharacteristic of me = 1 to Extremely characteristic of me = 5) that provides a global measure of aggression and four subscales: Physical Aggression (PA, 9 items), Verbal Aggression (VA, 5 items), Anger (A, 7 items), and Hostility (H, 8 items). The Buss–Perry Aggression Questionnaire-Short Form (BPAQ-SF; Bryant & Smith, 2001) is a “refined” version of the AQ consisting of 12 Likerttype items rated on a 6-point ordinal scale. The BPAQ-SF is also organized into four scales: Physical Aggression (PA, 3 items), Verbal Aggression (VA, items), Anger (A, 3 items), and Hostility (H, 3 items). Bryant and Smith (2001) decided to change the original 5-point scale to a 6-point scale to eliminate the scale's midpoint and force respondents to decide whether each statement was characteristic of them. Our Portuguese version consisted of Likert-type items rated on a 5-point ordinal scale (Never = 0 to Always = 4), restoring the scale's mid-point which was eliminated in Bryant and Smith's version, because five response alternatives with a neutral point seem to be more adequate in the context of cross-cultural comparisons and the reliability of a personality measurement instrument does not vary significantly when one response category is reduced (Gallardo-Pujol, Kramp, García-Forero, Pérez-Ramírez, & Andrés-Pueyo, 2006). The Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1989) is a brief self-report measure that evaluates self-esteem in adolescents and adults. The RSES can be scored by simply adding the ten items on a 4-point ordinal scale (Strongly Disagree = 0, Disagree = 1, Agree = 2, Strongly Agree = 3) after reverse scoring designated items (namely, items 2, 5, 6, 8 and 9). Higher scores indicate higher levels of selfesteem. A Portuguese version of the RSES was used (Pechorro, 2011). Pechorro, Marôco, Poiares, and Vieira (2011) found psychometric properties that justify the use of the RSES with Portuguese adolescent community and forensic populations, namely, in terms of internal consistency (Cronbach's α = .79), unidimensional factor structure (35.55% of variance), temporal stability (rs = .86; p ≤ .01), known-groups validity (Λ Wilks = .961; χ2 = 29.806; p ≤ .001), discriminant validity (r = .10; ns), corrected item–total correlation (range = .27–.62.) and average inter-item correlation (.27). Internal consistency for the present study, estimated by Cronbach's alpha, was .77. The Reactive–Proactive Aggression Questionnaire (RPQ; Raine et al., 2006) is a self-report measure that distinguishes between reactive and proactive aggression. The scale consists of 23 items rated on a 3-point ordinal scale (Never = 0, Sometimes = 1, Often = 2). A total of 11 items assess reactive aggression (e.g., “Reacted angrily when provoked by others”) and 12 items assess proactive aggression (e.g., “Hurt others to win a game”). Summed scores provide a measures of reactive or proactive aggression, as well as total aggression. Higher scores indicate higher levels of aggression. The RPQ is appropriate for use with youth in late adolescence and young adults. Internal consistency for adolescents has previously been reported as .86 for proactive aggression, .84 for reactive aggression, and .90 for total aggression (Raine et al., 2006). The Portuguese form of the RPQ was used (Pechorro, Ray, Raine, Maroco, & Gonçalves, in press) in this study. The internal consistency for the present study, estimated by Cronbach's alpha, was as follows: RPQ Reactive = .81; RPQ Proactive = .83; RPQ Total = .86. In addition, a questionnaire was constructed to describe the sociodemographic and criminal characteristics of the participants and to analyze the possible moderating effect of these variables. This questionnaire included variables such as participants' age, nationality, ethnic group, origin (rural vs. urban), level of schooling completed, socioeconomic status, parental marital status, number of siblings/half-siblings, previous use of physical violence, drug use and alcohol abuse. Socioeconomic status was measured by considering both parental level of education and profession, appropriate to the Portuguese reality (Simões, 1994).
2.2. Measures
2.3. Procedures
The Buss–Perry Aggression Questionnaire (BPAQ; Buss & Perry, 1992) is a four-factor model questionnaire consisting of 29 items scored
Appropriate procedures were followed during the translation and retroversion (Hambleton, Merenda, & Spielberger, 2005) of the
specificity required of this type of instrument to serve as a measurement model. Using confirmatory factor analysis (CFA) they found that Buss and Perry's original four-factor model achieved only mediocre fit in their three samples (Goodness of Fit Index [GFI] = .76–.81; χ2/df = 2.4–2.8). This state of affairs inspired Bryant and Smith (2001) to develop a reduced version of the BPAQ using five independent data sets. These authors removed all the items that loaded less than .40 or had crossed loadings (i.e., more than .40 in two or more components). This produced a shortened questionnaire consisting of three items for each factor. This modification resulted in a good fit to the four-factor model in several samples and explained an acceptable proportion of variance in both absolute and relative terms (GFI = .94). The reliabilities of the refined factors were corrected for differences in the number of constituent items (adjusted α's = .88–.92) using the Spearman–Brown prophecy formula, and were considered good. The data also provided strong support for the convergent and discriminant validity of the refined Physical Aggression, Anger, and Hostility factors, while the Verbal Aggression factor, showed poor discriminant validity. The authors concluded that their 12-item refined version (often referred to as BPAQ-SF) demonstrated superior psychometric properties in its overall goodness-of-fit to the data, yet still retained the conceptual framework originally proposed. The BPAQ and its short forms (Bryant & Smith, 2001; Buss & Perry, 1992; Buss & Warren, 2000) have proven their value in studying aggression profiles and predicting violent behavior in adults (e.g., Bushman & Wells, 1998; Diamond & Magaletta, 2006; Diamond, Wang, & Buffington-Vollum, 2005), while the same cannot be said about adolescents and preadolescents. The trajectory of developmental aggression (e.g., low, persistent or desisting aggression) among youth (e.g. Collishaw, Maughan, Goodman, & Pickles, 2004; Martino, Ellickson, Klein, McCaffrey, & Edelen, 2008), from late childhood through adolescence, needs valid measurement. The adaptation of the BPAQ-SF to youth within this age range would be very useful in terms of obtaining a short and valid measurement of the aggression trait. The main aim of the present study is to validate a Portuguese version of the Buss–Perry Aggression Questionnaire-Short Form (BPAQ-SF; Bryant & Smith, 2001) so as to further the investigation of aggression among juvenile delinquents and community youths in Portugal and Portuguese speaking countries. 2. Method 2.1. Participants
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx
instrument. The initial translation from English into Portuguese was completed by the first and second authors of this article, who made sure that young people would be able to properly understand the meaning of the items. The questionnaire was then independently translated back into English by a native English speaker with considerable professional experience in translating psychology-related scientific texts. No significant differences were found between the back-translation and the original version, demonstrating that the translated items had the same or very similar meanings as the original English items. The original items and the Portuguese translation are available in Appendix A and Appendix B, respectively. Authorization to assess youths was obtained from the General Directorate of Reintegration and Prison Services, Ministry of Justice (DireçãoGeral de Reinserção e Serviços Prisionais, Ministério da Justiça). The detainees from the existing male Juvenile Detention Centers were informed about the nature of the study and asked to participate. The measures were administered by means of individual face-to-face interviews in an appropriate setting. The majority of the participants were convicted of serious and violent crimes (e.g., homicide, robbery, assault, rape, child sexual abuse), and were detained by the court's decision. Authorization was also obtained from the Directorate General of Education, Ministry of Education (Direção-Geral de Educação, Ministério da Educação). The community sample was recruited from schools in the Lisbon region (Portugal) that agreed to participate. Written consent was required from each participating student's parent or guardian. Participants who were unwilling or unable to collaborate were excluded. The data was analyzed using SPSS v22 (IBM SPSS, 2013) and EQS 6.2 (Bentler & Wu, 2008). Following data entry 15% of the questionnaires were randomly selected to evaluate accuracy of data input, which was judged to be good. The factor structure of the Portuguese language version of the BPAQ was assessed with Confirmatory Factor Analysis (CFA) performed in EQS 6.2 (Bentler & Wu, 2008; Byrne, 2006), with robust estimation methods appropriate for ordinal variables and evaluation of model fit. The goodness of fit indices calculated included chi-square/ degrees of freedom, Satorra–Bentler chi-square/degrees of freedom, comparative fit index (CFI) and root mean square error of approximation (RMSEA). A chi-square/degrees of freedom value b 5 is considered adequate, ≤2 is considered good and 1 = very good (Marôco, 2014; West, Taylor, & Wu, 2012). A CFI ≥ .90 and RMSEA ≤ .10 indicate adequate fit, whereas a CFI ≥ .95 and RMSEA ≤ .06 indicate good model fit (Byrne, 2006). Items with standardized loading above .45 were retained. The chi-square test was used to compare nominal variables. Pearson's r was used to analyze correlations between scale variables. Point-bisserial correlations were used to analyze the association between nominal dichotomous variables and scale variables. Discriminant analysis was used to assess known-groups validity. Results were considered significant if p ≤ .05 and marginally significant if p ≤ .1 (Aron, Coups, & Aron, 2013).
3. Results The aim of our study was to validate a short version of the Buss– Perry Aggression Questionnaire for Portuguese youths. Our first step in attempting to validate the Portuguese version of the BPAQ-SF was to try to confirm the four-factor structure obtained in previous studies carried out using the EQS 6.2 software (Bentler & Wu, 2008). No support was found for the original BPAQ by means of CFA using the ML method due to poor fits and/or low item loadings. The BPAQ-SF four-factor structure showed a better fit, but using EQS's ML Robust method for skewed data we were able to obtain the best fit for the BPAQ-SF (see Table 1). In Fig. 1 are the loadings for the four-factor structure of the BPAQ-SF estimated with the ML Robust method. Set in Table 2 are the correlation matrix values.
3
Table 1 Goodness of fit indexes for the different models.
BPAQ 4-factor BPAQ-SF 4-factor BPAQ-SF 4-factor Robust
χ2/df or S-Bχ2/df
CFI
RMSEA (90% CI)
2.86 2.94 1.43
.80 .97 .99
.07 (.06, .07) .07 (.06, .08) .03 (.01, .05)
Note. BPAQ = Aggression Questionnaire; BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; χ2 = chi-square; S-Bχ2 = Satorra–Bentler chi-square; df = degrees of freedom; CFI = comparative fit index; RMSEA (90% CI) = root mean square error of approximation (90% confidence interval); ML = maximum likelihood.
The following step was the estimation of Cronbach's alpha, mean inter-item correlation and corrected item–total correlation range (see Table 3). The convergent validity of the BPAQ-SF with the RPQ and its dimensions revealed a statistically significant positive correlation, while the discriminant validity with the RSES revealed a statistically significant negative correlation (see Table 4). The predictive validity of the BPAQ-SF score with previous physical violence (coded No = 0, Yes = 1) revealed the existence of a statistically significant positive point-bisserial correlation. Similarly, the associations of the BPAQ-SF with criterion-related variables such as drug use (coded No = 0, Yes = 1) and alcohol abuse (coded No = 0, Yes = 1) also revealed the existence of statistical significant positive pointbisserial correlations (see Table 5). The known-groups validity of the BPAQ-SF was calculated using Wilks' lambda; statistically significant differences were found between the forensic sample and the community sample (Wilks' λ = .96, χ2 = 5.147 (1), p ≤ .05). 4. Discussion The present study had as its aim the analysis of the psychometric properties of the BPAQ-SF, a shortened and refined version of the measure that is considered the gold standard in the assessment of aggression, among Portuguese youth. In terms of factor structure, no support was found for the original BPAQ by means of CFA using the ML method due to poor fits. This was somewhat expected because some of the items had non-normal distributions. The BPAQ-SF showed an adequate fit using the same ML method. However, by using the ML Robust method developed for skewed data, we were able to obtain a very good fit confirming the four-factor structure of the BPAQ-SF. Evidence obtained in our study shows that Bryant and Smith's (2001) refined four-factor model fits our data very well. Some psychometric problems were found when analyzing our Portuguese adaptation of the BPAQ-SF. Analysis of the internal consistency revealed somewhat low values for the Verbal Aggression (VA) and the Anger (A) factors that may put into question the reliability of measurement for these two factors (Cortina, 1993; Kaplan & Saccuzzo, 2009). These values were lower than the reliability estimates obtained by Bryant and Smith (2001), which ranged from .79 to .80 for physical aggression, .73 to .83 for verbal aggression, .71 to .76 for anger, and .70 to .75 for hostility. Because Cronbach's alpha is sensitive to the number of items included in a given scale, these authors also calculated adjusted alpha coefficients of .88 to .92 for the four subscales. Alpha values for an overall aggressiveness score were not provided. There were also some positive findings. Regarding the mean interitem correlations (which is a straightforward measure of internal consistency), values were within the recommended range of .15–.50 (Clark & Watson, 1995; Domino & Domino, 2006). In terms of the corrected item–total correlation range, the BPAQ-SF total score and its four factors all reached the minimum recommended value of .20 (Kaplan & Saccuzzo, 2009; Nunnally & Bernstein, 1994). The convergent validity of the BPAQ-SF with the RPQ and its two factors revealed a moderate–high to strong statistically significant correlation, while the discriminant validity with the RSES revealed a
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
4
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx
Fig. 1. BPAQ-SF four-factor ML Robust structure with standardized item loadings.
moderate–low negative correlation, demonstrating the expected construct overlap of the former but not of the latter (American Psychological
Table 2 Correlation matrix.
1 PA 2 VA 3A 4H 5 BPAQ-SF
1 BPAQ-SF PA
2 BPAQ-SF VA
1 .88⁎ .79⁎ .50⁎ .79⁎
1 .94⁎ .45⁎ .78⁎
3 BPAQ-SF A
4 BPAQ-SF H
5 BPAQ-SF total
Association, 1999; Kaplan & Saccuzzo, 2009). This is in line with those found in previous studies (Bryant & Smith, 2001; Williams et al., 1996). As expected (American Psychological Association, 1999), and in keeping with previous studies (e.g. Diamond & Magaletta, 2006), a Table 3 Cronbach's alpha, mean inter-item correlation, and corrected item–total correlation range. BPAQ-SF PA BPAQ-SF VA BPAQ-SF A BPAQ-SF H BPAQ-SF total
1 .64⁎ .81⁎
1 .70⁎
Cronbach's α .71 MIIC .45 CITCR .49–.55 1
Note. BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; BPAQ-SF PA = Physical Aggression dimension; BPAQ-SF VA = Verbal Aggression dimension; BPAQ-SF A = Anger dimension; BPAQ-SF H = Hostility dimension ⁎ Significant at the .05 level.
.63 .36 .37–.48
.62 .36 .37–.47
.75 .49 .50–.63
.84 .30 .40–.59
Note. BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; BPAQ-SF PA = Physical Aggression dimension; BPAQ-SF VA = Verbal Aggression dimension; BPAQ-SF A = Anger dimension; BPAQ-SF H = Hostility dimension; Cronbach's α = Cronbach's alpha; MIIC = mean inter-item correlation; CITCR = corrected item–total correlation range.
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx Table 4 Convergent validity with RPQ Reactive, RPQ Proactive and RPQ Total, and discriminant validity with RSES.
BPAQ-SF
RPQ Reactive
RPQ Proactive
RPQ Total
RSES
r = .61⁎⁎⁎
r = .51⁎⁎⁎
r = .65⁎⁎⁎
r = −.38⁎⁎⁎
Note. BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; RPQ = Reactive–Proactive Aggression Questionnaire; RPQ Proactive = Proactive dimension; RPQ Reactive = Reactive dimension; RSES = Rosenberg Self-Esteem Scale. ⁎⁎⁎ Significant at the .001 level.
Table 5 Correlations of BPAQ-SF with previous use of physical violence, drug use, and alcohol abuse.
BPAQ-SF
PPV
DU
AB
rpb = .25⁎⁎
rpb = .20⁎
rpb = .17⁎
Note. BPAQ-SF = Buss–Perry Aggression Questionnaire-Short Form; PPV = previous physical violence; DU = drug use; AB = alcohol abuse; rpb = point-bisserial correlation. ⁎⁎ Significant at the .01 level. ⁎ Significant at the .05 level.
5
Appendix A Original items of the BPAQ-SF. Items 1 (2). Given enough provocation, I may hit another person. 2 (6). There are people who pushed me so far that we came to blows. 3 (8). I have threatened people I know. 4 (11). I often find myself disagreeing with people. 5 (13). I can’t help getting into arguments when people disagree with me. 6 (14). My friends say that I’m somewhat argumentative. 7 (15). I flare up quickly but get over it quickly. 8 (20). Sometimes I fly off the handle for no good reason. 9 (21). I have trouble controlling my temper. 10 (23). At times I feel I have gotten a raw deal out of life. 11 (24). Other people always seem to get the breaks. 12 (25). I wonder why sometimes I feel so bitter about things.
Appendix B Portuguese translation of the BPAQ-SF. Items
moderate positive correlation was found between the predictive validity of the BPAQ-SF score with a history of previous physical violence. The correlations of the BPAQ-SF with criterion-related variables revealed the existence of moderate–low associations with drug use and alcohol abuse. Associations between aggression scores and drug use and alcohol abuse have been consistently reported in the literature (e.g., McMurran et al., 2006; Taylor & Hulsizer, 1998), although the ones we obtained were somewhat lower but still statistically significant. Also consistent with prior studies (e.g. Smith & Waterman, 2006), the known-groups validity revealed that the BPAQ-SF could significantly discriminate between the forensic group and the community group, conceptualized as structurally different and mutually exclusive (Kaplan & Saccuzzo, 2009). Overall, our findings provide some additional support for Buss and Perry's (1992) four-factor model of aggression among adolescents and its potential generalization across different languages and cultures. It was possible to demonstrate some appropriate psychometric properties that justify the future use of the BPAQ-SF with the Portuguese youth population. To our knowledge this is the first study attempting the validation of the BPAQ-SF among juvenile delinquents in Portugal. Additional validation procedures are recommended and should be done in the future (e.g., test–retest reliability, cross-validation using other samples). Given the enormous costs that violent offenders create in terms of their crimes on victims and society as a whole, we hope our study may guide the use of the BPAQ-SF with Portuguese youth and contribute to the betterment of treatment programs targeting serious and violent juvenile offenders in Portugal.
Acknowledgments We wish to thank the staffs of the following Portuguese juvenile detention centers for their collaboration in this study: Belavista, Mondego, Navarro de Paiva, Olivais, Padre António Oliveira, Santo António, Santa Clara, and Prisão-Escola de Leiria. This study was partly supported by the Portuguese Foundation for Science and Technology (FCT; Grant SFRH/BPD/86666/2012) with co-financing of the European Social Fund (POPH/FSE), the Portuguese Ministry of Education and Science (MEC) through national funds, and co-financed by FEDER under the PT2020 Partnership Agreement (UID/PSI/01662/2013).
1 (2). Se me provocarem bastante, posso bater noutra pessoa. 2 (6). Houve pessoas que me pressionaram tanto que andámos à pancada (luta). 3 (8). Já ameacei pessoas que conheço. 4 (11). Entro em desacordo com as pessoas. 5 (13). Não consigo deixar de discutir quando as pessoas discordam de mim. 6 (14). Os meus amigos dizem que gosto de discutir. 7 (15). Exalto-me facilmente, mas passa-me rapidamente. 8 (20). Perco o controlo sem razão em especial. 9 (21). Tenho dificuldade em controlar o meu feitio. 10 (23). Sinto que a vida não me dá boas oportunidades. 11 (24). As outras pessoas parecem ter sempre as melhores oportunidades. 12 (25). Pergunto-me por que motivo às vezes me sinto tão amargo com as coisas.
References American Psychological Association (1999). Standards for educational and psychological tests. Washington, DC: Author. Anderson, C., & Bushman, B. (2002). Human aggression. Annual Review of Psychology, 53, 27–51. Andreu, J., Peña, M., & Graña, J. (2002). Adaptación psicométrica de la versión española del cuestionario de agresión [Psychometric adaptation of the Spanish version of the Aggression Questionnaire]. Psicothema, 14, 476–482. Archer, J., Kilpatrick, G., & Bramwell, R. (1995). Comparison of two aggression inventories. Aggressive Behavior, 21, 371–380. Aron, A., Coups, E., & Aron, E. (2013). Statistics for psychology (6th ed.). New York, NY: Pearson. Bentler, P., & Wu, E. (2008). EQS for Windows user's guide. Encino, CA: Multivariate Software, Inc. Berkowitz, L. (1993). Aggression: Its causes, consequences and control. Philadelphia: Temple University Press. Bryant, F., & Smith, B. (2001). Refining the architecture of aggression: A measurement model for the Buss–Perry Aggression Questionnaire. Journal of Research in Personality, 35, 138–167. Bushman, B., & Wells, G. (1998). Trait aggressiveness and hockey penalties: Predicting hot tempers on the ice. Journal of Personality and Social Psychology, 83, 969–974. Buss, A. (1961). The psychology of aggression. New York: John Wiley & Sons, Inc. Buss, A., & Perry, M. (1992). The Aggression Questionnaire. Journal of Personality and Social Psychology, 63(3), 452–459. Buss, A., & Warren, W. (2000). The Aggression Questionnaire manual. Los Angeles: Western Psychological Services. Byrne, B. (2006). Structural equation modeling with EQS: Basic concepts, applications, and programming. Mahwah, NJ: Lawrence Erlbaum Associates. Cima, M., & Raine, A. (2009). Distinct characteristics of psychopathy relate to different subtypes of aggression. Personality and Individual Differences, 47, 835–840. Clark, L., & Watson, D. (1995). Constructing validity: Basic issues in objective scale development. Psychological Assessment, 7(3), 309–319. Collishaw, S., Maughan, B., Goodman, R., & Pickles, A. (2004). Time trends in adolescent mental health. Journal of Child Psychology & Psychiatry, 45, 1350–1362. Cortina, J. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78(1), 98–104. Cunha, O., & Gonçalves, R. (2012). Análise confirmatória fatorial de uma versão portuguesa do Questionário de Agressividade de Buss–Perry [Confirmatory factor analysis of a Portuguese version of the Buss–Perry Aggression Questionnaire]. Laboratório de Psicologia, 10(1), 3–17.
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033
6
P. Pechorro et al. / International Journal of Law and Psychiatry xxx (2015) xxx–xxx
Diamond, P., & Magaletta, P. (2006). The Short-Form Buss–Perry Aggression Questionnaire (BPAQ-SF): A validation study with federal offenders. Assessment, 13(3), 227–240. Diamond, P., Wang, E., & Buffington-Vollum, J. (2005). Factor structure of the Buss–Perry Aggression Questionnaire (BPAQ) with mentally ill male prisoners. Criminal Justice and Behavior, 32(5), 546–564. Domino, G., & Domino, M. (2006). Psychological testing: An introduction (2nd ed.). New York: Cambridge University Press. Fossati, A., Maffei, C., Acquarini, E., & Di Ceglie, A. (2003). Multigroup confirmatory component and factor analyses of the Italian version of the Aggression Questionnaire. European Journal of Psychological Assessment, 19, 54–65. Gallardo-Pujol, D., Kramp, U., García-Forero, C., Pérez-Ramírez, M., & Andrés-Pueyo, A. (2006). Assessing aggressiveness quickly and efficiently: The Spanish adaptation of Aggression Questionnaire-Refined version. European Psychiatry, 21(7), 487–494. Hambleton, R., Merenda, P., & Spielberger, C. (2005). Adapting educational and psychological tests for cross-cultural assessment. Mahwah, NJ: Lawrence Erlbaum Associates. Harris, J. (1995). Confirmatory factor analysis of the aggression questionnaire. Behavioral Research and Therapy, 8, 991–993. Harris, J. A. (1997). A further evaluation of the Aggression Questionnaire: Issues of validity and reliability. Behavior Research and Therapy, 35, 1047–1053. IBM SPSS (2013). IBM SPSS statistics base 22. Chicago, IL: SPSS. Kaplan, R., & Saccuzzo, D. (2009). Psychological testing: Principles, applications, and issues. Belmont, CA: Wadsworth, Cengage Learning. Little, T., Jones, S., Henrich, C., & Hawley, P. (2003). Disentangling the ‘whys’ from the ‘whats’ of aggressive behaviour. International Journal of Behavioral Development, 27, 122–133. Lovas, L., & Trenkova, S. (1996). Aggression perception of an incident. Studia Psychologia, 38, 265–270. Marôco, J. (2014). Análise de Equações Estruturais: Fundamentos teóricos, software & aplicações [Structural equations analysis: Theoretical foundations, software and applications]. Pero Pinheiro: ReportNumber, Ltd. Martino, S., Ellickson, P., Klein, D., McCaffrey, D., & Edelen, M. (2008). Multiple trajectories of physical aggression among adolescent boys and girls. Aggressive Behavior, 34, 61–75. Maxwell, J. (2007). Development and preliminary validation of a Chinese version of the Buss–Perry Aggression Questionnaire in a population of Hong Kong Chinese. Journal of Personality Assessment, 88(3), 284–294. McMurran, M., Egan, V., Cusens, B., Van den Bree, M., Austin, E., & Charlesworth, P. (2006). The alcohol-related Aggression Questionnaire. Addiction Research and Theory, 14(3), 323–343. Meesters, C., Muris, P., Bosma, H., Schouten, E., & Beuving, S. (1996). Psychometric evaluation of the Dutch version of the Aggression Questionnaire. Behavioral Research Therapy, 34, 839–843. Mejovsek, M., Budanovac, A., & Sucur, Z. (2000). The relationship between inmates' aggression and their socioeconomic and family characteristics. Hrvatska Revija Za Rehabilitacijska Istrazivanja, 36, 75–86. Morren, M., & Meesters, C. (2002). Validation of the Dutch version of the Aggression Questionnaire in adolescent male offenders. Aggressive Behavior, 28, 87–96. Nahama, V., Ayoub, M. P., Borie, R., & Petit, F. (2003). Problemes d’adaptation sociale chez des jeunes de 9 a 12 ans presentant des troubles des apprentissages [Social adjustment problems of youngsters with learning difficulties]. Approche Neuropsychologique des Apprentissages chez l’Enfant, 15, 9–13. Nakano, K. (2001). Psychometric evaluation on the Japanese adaptation of the Aggression Questionnaire. Behavioral Research Therapy, 39, 853–858. Nunnally, J., & Bernstein, I. (1994). Psychometric theory. New York, NY: McGraw-Hill.
Parrott, D., & Giancola, P. (2007). Addressing “The criterion problem” in the assessment of aggressive behavior: Development of a new taxonomic system. Aggression and Violent Behavior, 12, 280–299. Pechorro, P. (2011). Juvenile delinquency: Study of some psychological and relational variables with an emphasis on psychopathic traits (Unpublished doctoral dissertation). Faculdade de Medicina da Universidade de Lisboa, Lisbon, Portugal. Pechorro, P., Marôco, J., Poiares, C., & Vieira, R. (2011). Validação da Escala de Auto-Estima de Rosenberg com adolescentes portugueses em contexto forense e escolar [Validation of the Rosenberg Self-Esteem Scale with Portuguese adolescents in forensic and school contexts]. Arquivos de Medicina, 25(5/6), 174–179. Pechorro, P., Ray, J., Raine, A., Maroco, J., & Gonçalves, R. (2015). The Reactive–Proactive Aggression Questionnaire: Validation among a Portuguese sample of incarcerated juvenile delinquents. Journal of Interpersonal Violence, http://dx.doi.org/10.1177/ 0886260515590784 (in press). Prochazka, H., & Agren, H. (2001). Aggression in the Swedish population measured with a new self-rating inventory: The Aggression Questionnaire-revised Swedish version (AQ-RSV). Nordic Journal of Psychiatry, 55, 17–24. Raine, A., Dodge, K., Loeber, R., Gatzke-Kopp, L., Lynam, D. R., Reynolds, C., et al. (2006). The reactive-proactive aggression questionnaire: Differential correlates of reactive and proactive aggression in adolescent boys. Aggressive Behavior, 32, 159–172. Rosenberg, M. (1989). Society and the adolescent self-image (Rev. ed.). Middletown, CT: Wesleyan University Press. Ruchkin, V., & Eisemann, M. (2000). Aggression and psychological problems in juvenile male delinquents versus controls in Russia: Alternate ways of ‘letting off steam’? Aggression and Violent Behavior, 5, 217–225. Santisteban, C., Alvarado, J., & Recio, P. (2007). Evaluation of a Spanish version of the Buss and Perry aggression questionnaire: Some personal and situational factors related to the aggression scores of young subjects. Personality and Individual Differences, 42, 1453–1465. Simões, A. (1993). São os homens mais agressivos que as mulheres? [Are men more aggressive than women?]. Revista Portuguesa de Pedagogia, XXVII(3), 387–404. Simões, M. (1994). Investigação no âmbito da aferição nacional ao Teste das Matrizes Progressivas Coloridas de Raven [Researches regarding the national validation of Raven's Coloured Progressive Matrices Test] (Unpublished doctoral dissertation). Faculdade de Psicologia e Ciências da Educação da Universidade de Coimbra, Coimbra, Portugal. Smith, P., & Waterman, M. (2006). Self-reported aggression and impulsivity in forensic and non-forensic populations. Journal of Family Violence, 21, 425–437. Taylor, S., & Hulsizer, M. (1998). Psychoactive drugs and human aggression. In R. Geen, & E. Donnerstein (Eds.), Human aggression: Theories, research, and implications for social policy (pp. 139–165). New York: Academic Press. Vigil-Colet, A., Lorenzo-Seva, U., Codorniu-Raga, M., & Morales, F. (2005). Factor structure of the Buss–Perry Aggression Questionnaire in different samples and languages. Aggressive Behavior, 31, 601–608. Von Collani, G., & Werner, R. (2005). Self-related and motivational constructs as determinants of aggression. An analysis and validation of a German version of the Buss–Perry aggression questionnaire. Personality and Individual Differences, 38, 1631–1643. West, S., Taylor, A., & Wu, W. (2012). Model fit and model selection in structural equation modeling. In R. Hoyle (Ed.), Handbook of structural equation modeling (pp. 209–231). New York, NY: Guilford Press. Williams, T., Boyd, J., Cascardi, M., & Poythress, N. (1996). Factor structure and convergent validity of the aggression questionnaire in an offender population. Psychological Assessment, 4, 398–403.
Please cite this article as: Pechorro, P., et al., Validation of the Buss–Perry Aggression Questionnaire-Short Form among Portuguese juvenile delinquents, International Journal of Law and Psychiatry (2015), http://dx.doi.org/10.1016/j.ijlp.2015.08.033