This exploratory study was designed to use the attributional model to explore adolescents' perceptions about the causes of peer suicides. Additionally, variables related to the adolescent's family background, social relationships and self were examined to determine factors that may influence suicidal behaviors.

Four hundred and eighty-six eleventh and twelfth grade students from a suburban public school district near a large metropolitan area in the Northeast completed three self-report measures, an open-ended question measuring causality, a scenario four-point scale rating blame, and a background questionnaire. Forty-two students were seen in follow-up interviews to verify data.

Two major tenets of attribution theory are that people interpret events and the consequences of the events to arrive at decisions about their personal dispositions, and that a person makes different attributions depending on whether he or she is in the actor or observer position.

Data were analyzed using analyses of variance and crosstabulations for three groups, male and females, suicidal behaviors and knowledge of a suicidal person. Specific tests included the Dunn Multiple Range Test, Fisher's Exact Test (Two-Tailed) and chi-square.

The findings indicate that 23% of the high school students studied, reported self hurt behaviors and 6.7% reported suicide attempts. A theme of "too much pressure" was reported by 40% of the adolescents as a cause of suicide. The quality of family life and parental acts of affection were identified as very important ways to decrease suicidal behaviors. Significant differences were seen in internal/external causality (p $<$.05) for sex differences and suicidal behaviors, depending on the test measure used. Divergence from anticipated findings indicated that adolescents with suicidal behaviors (actors) made fewer external attributions than the non-suicidal (observer) adolescents. Parent's marital status, church attendance, school performance/attendance, social supports, knowledge of a suicidal person and health were statistically significant when students were compared by suicidal behaviors and suicidal ideation.

Results were discussed in terms of the need for additional research with adolescent populations and health professionals.