Bullying is an important risk factor for mental health of children and adolescents. Bullying is a subset of aggressive behavior, generally defined as an act intended to inflict injury or discomfort upon another student. There is considerable evidence for emotional problems (such as, depression, anxiety, and low self-esteem) among children and young people who experience bullying or peer victimization, and between bullying and behavioral problems (aggression, antisocial behavior, and substance use), but cross-sectional studies cannot explore the direction of the relationship. Besides, longitudinal studies are crucial to explore stability of being victim and being aggressor. There are no many studies that has attempted to do this. In Latin American countries, such as Argentina, longitudinal studies are very scarce in this respect. Exploring the direction of the link between victimization and emotional problems is vital to identify targets of victimization in adolescents and reducing cost of victimization. So, the purpose of this research was examining stability of victims and aggressor in three-time points and determine the direction of the relationship between emotional problems (depression and anxiety) and victimization, on one hand, and between behavioral problems (antisocial behavior and aggression), on the other hand. To this end, data were obtained from a sample of 485 students from two secondary schools attending first course, second and third in Parana, Entre Rios. A 58% was female and 42% was males; Mage = 13.6. They were examined in May (Time 1, T1), August (Time 2, T2), and November (Time 3, T3). To measure victimization, they completed Olweus Bully/ Victim Questionnaire (tapping verbal, physical, relational, and cybervictimization and these kind of bullying), Kovacs Child Depression Inventory for depression, Rosenberg Psychosomatic Symptom Scale for anxiety, and aggressive behavior and antisocial behavior from Achebanch and Rescorla ASEBA (2001), and some sociodemographic questions. Confidentiality and anonymity were ensured. Date was set up with Statistical Package for Social Sciences version 22 and AMOS 16 to set up structural equation modeling (SEM). Results suggested that 15%,16%, and 18% were victims in T1, T2, and T3. As regarding being aggressor, the percentages were 9%, 10%, and 9%. With respect to stability of victim status in each of the three times, a 39% was victim in the three times; as regarding aggressor status, a 57% was aggressor in the three times. An autoregressive cross-lagged with index of victimization, scores on depression and anxiety in T1, T2, and T3 indicated an adequate model fit: CFI = .99, TLI = .96 and RMSEA =.08. Victimization at T1 predicted victimization at T2 and T3. Regarding the links between victimization and emotional problems, this relationship was reciprocal with respect to depression: victimization predicted depression at T2 and T3 and depression predicted victimization in both times, as well. The links between victimization and anxiety was non-significant. Reciprocal links between depression and anxiety were found. With respect to bullying, antisocial behavioral and aggression model, results indicated a good model fit: CFI = .99, TLI = .96, and RMSEA = .06. Findings suggested reciprocal links between bullying and aggression. Links between bullying and antisocial behavioral were non-significant. These results suggested stability in victim status and aggressor status, but also change in only a period of six month. Besides, results showed that the link between victimization and depression was reciprocal: depression was an antecedent but a consequence of victimization, as well; the same happened for bullying and aggression. So, these results are in line with a vast literature suggesting reciprocal effects. In the discussion, we provided further explanation of the results, considering domination theory among others. Finally, limitations and suggestion for carrying out future studies are provided.
El objetivo de este trabajo fue explorar la estabilidad del ser víctima y ser agresor en tres puntos en el tiempo y la direccionalidad de la causalidad entre la victimización y los problemas emocionales, por un lado, y entre la agresión y los problemas de conducta en adolescentes, por el otro. Se constituyó una muestra intencional de 485 alumnos de dos escuelas secundarias de Paraná, Entre Ríos. Un 42% correspondió a alumnos varones; la media de edad fue de 13.6 años. Los participantes contestaron un cuestionario sociodemográfico, el Cuestionario de Agresores/Víctimas de Olweus (que mide bullying verbal, físico, relacional y cyberbullying), el Inventario de Depresión de Kovacs, la Escala de Síntomas Psicosomáticos de Rosenberg para ansiedad y las escalas de conducta antisocial y agresividad del ASEBA de Achenbach y Rescorla (2001). Los resultados indicaron que un porcentaje similar era víctima en cada uno de los tres tiempos (15%, 16% y 18%). Con respecto al ser agresor, también se hallaba un porcentaje similar en los tres tiempos (9% 10% y 9%). Considerando el grupo que era victimizado en el tiempo 1, un 39% lo era también en los tiempos 2 y 3, mientras que dentro del 9% que era agresor en el tiempo 1, un 57% lo era en los dos tiempos subsiguientes. Un modelo autoregressive cross-lagged indicó que existían efectos bidireccionales entre la victimización y la depresión. También se hallaron efectos bidireccionales para el bullying y la agresividad. Estos resultados indican estabilidad, pero también cambio en el ser victimizado y ser agresor. En la discusión se analizan estos hallazgos a partir de diversos aportes teóricos para explicar tanto la estabilidad como el cambio. Finalmente, se proveen limitaciones y sugerencias para futuros estudios.