Rough-and-tumble play to cope with physical aggression in particular settings: an exploratory study with young adolescents

Background Despite the fact that aggressive behaviors in young adolescents have received considerable attention, the research for strategies aimed at reducing aggressiveness is still a crucial question. Rough-and-tumble play (R&T) could provide a functional-adaptive account to cope with aggressive behaviors (Carraro, Mauro & Ventura, 2006; Steyn & Roux, 2009; Carraro, Gobbi & Moè, submitted). The purpose of the present study was to analyze the effect of R&T in reducing aggressiveness in a group of young adolescents with behavioral problems. Method The study was carried out in 3 community centers (CC1, CC2 & CC3) for young adolescents. In order to be admitted to a CC, adolescents must be indicated by school teachers as having behavioral problems. Fifty-four young adolescents (CC1=23, CC2=15, CC3=16; mean age=13.3±1.01) were invited to take part in the study. Some of them dropped out or refused to take part in the activities. In the end, 37 adolescents completed the study (CC1=19, CC2=9, CC3=9). Seven participants were Italian, the others were from immigrant backgrounds. This second group exhibited considerable difficulties with the Italian language. Participants filled in the short version of the Buss & Perry Aggression Questionnaire (AQ12; Bryant & Smith, 2001) both pre- and post-intervention, which includes 4 subscales: Physical Aggressiveness (PA), Verbal Aggressiveness (VA), Anger (AN) and Hostility (HO). In CC1 participants were invited to take part in R&T, consisting in games based on chasing and wrestling. Participation was on a voluntary basis, so 8 adolescents participated, while 11 spent their time in free play. In CC2 adolescents were required to participate in role playing and discussions aimed at reducing aggressiveness, which required no physical contact. All activities were proposed once a week for 6 consecutive weeks. In CC3 no extra activities were proposed. Adolescents spent their time in free play. Results Cronbach’s Alpha was acceptable for the PA and HO subscales (62 and .68 respectively) and not acceptable for VA (.45) and AN (.59). We decided to analyze the PA subscale due to its stronger link to R&T play. Independent sample t-test revealed no significant differences in the 4 subscales at the pre-test between the CCs. Paired sample t-test did not report any significant variation. However, between pre- and post-test a reduction in PA was observed in the adolescents of CC1 attending R&T (-14.2%; p=.20), of CC2 (-17.0%; p=.31) and of CC3 (-16.3%; p=.08). Only adolescents attending CC1 but not involved in R&T increased their PA level (+7.5%; p=.28). Conclusion Results do not allow for a firm conclusion concerning the effect of a short-term R&T program or of other traditional activities in reducing self-perceived aggressiveness. Nonetheless, analysis of limitations can help to better plan further studies. Statistically significant results have been obtained with studies with a longer experimental time (Palermo, 2006). We may suppose that 6 classes are insufficient to have a significant impact on aggressiveness. The AQ12 reported low Cronbach’s Alphas in two subscales, probably because it was difficult for most of the foreign participants to fully understand the items. Moreover, the great number of adolescents who dropped out as well as low motivation to take part in proposed activities were limitations. In conclusion, it seems possible to manage R&T in a setting such as CCs. However, limitations linked to drop-out, evaluation instruments, motivation and length of intervention should be carefully considered in further studies.