The aims of this study were to know risk and protective factors for dissocial behavior keeping in mind that the self-report of dissocial behavior is biased by the impression management. A probability sample of adolescents that lived in two neighborhoods with high indexes of gangs and offenses (112 male and 86 women) was collected. The 27-item Dissocial Behavior Scale (ECODI27; Pacheco & Moral, 2010), Balanced Inventory of Desirable Responding, version 6 (BIDR-6; Paulhus, 1991), Sensation Seeking Scale, form V (SSS-V; Zuckerman, Eysenck, & Eysenck, 1978), Parent-Adolescent Communication Scale (PACS; Barnes & Olson, 1982), 30-item Rathus Assertiveness Schedule (RAS; Rathus, 1973), Interpersonal Reactivity Index (IRI; Davis, 1983) and a social relationship questionnaire (SRQ) were applied. Binary logistic regression was used for the data analysis. A third of the participants showed dissocial behavior. Belonging to a gang in the school (schooled adolescents) or to a gang out of school and job (total sample) and desinhibition were risk factors; being woman, perspective taking and open communication with the father were protective factors. School-leaving was a differential aspect. We insisted on the need of intervention on these variables.