Following an alarming rise in suicide rates during the past few decades, it now ranks among the major causes of death in the Province of Quebec (Charron, 1981). Rates for adolescents in particular have increased (National Task Force on Suicide in Canada, 1987). At present, about 15% of all fatalities in this age group are suicide related. The scope of the problem is even greater when suicide attempts and ideation are considered. Charron (1981) and Morisette (1984) estimated that for every completed suicide there are 4 to 10 unsuccessful attempts. Tousignant, Hamel, and Bastien (1988) found that 6.7% of surveyed high school students had tried to commit suicide at least once. Regarding ideation, Tousignant et al. (1988), Cote, Pronovost, and Ross (1989), and Tousignant, Hanigan, and Bergeron (1984) respectively reported that 13.2% and 15.4% of high school students and 21.2% of college students had seriously thought about committing suicide at some time during their lives. Emond et al. (1988) noted that the suicidal ideation rates for 15- to 24-year-olds are almost double those of other age groups.
Adolescent suicide has become a major concern, and many studies of the behavior have been conducted to date. Nevertheless, much controversy still surrounds its etiology, and specific risk factors for adolescent suicide have not yet been conclusively identified (Schwartz, Hoffman, & Riddile, 1987). Moreover, most previous research has not used multivariate analysis techniques or explored gender differences (Simons & Murphy, 1985).
Because suicidal ideation logically precedes suicidal acts, the identification of true predictors of ideation would permit a better understanding of suicidal risk. Therefore, the present study focused on selected suicide-related factors frequently reported in the literature in order to isolate, by multivariate techniques, those variables that are prominent predictors of suicidal ideation in adolescents.
Charron's (1981) suggestion that suicidal ideation is a function of personal characteristics, negative life experiences, and social support formed the theoretical basis for the selection of variables. The personal variables addressed in the present study were sex, age, self-esteem, locus of control, depression, and alcohol and other drug use. Negative life experiences included stress, perceived health status, parental control in child rearing, family status, physical abuse, academic performance, and school absenteeism. Regarding social support, the study addressed level of support, satisfaction with received support, and alienation.
Statistical data indicate that suicidal behavior varies with sex and age. Girls seem to be more prone to suicidal thoughts (Charron, 1981; Cote et al., 1989) and make attempts more often than do boys (Baker, 1983; Corbeil, 1984; Cote et al., 1989; Garfinkel, Froese, & Hook, 1982; Tousignant et al., 1988), whereas boys lead in number of completed suicides (Rosenthal, 1981; Sommer, 1984; Withers & Kaplan, 1987). Regarding age, it has been found that suicidal ideation is more prevalent among older adolescents (Emond et al., 1988), and that, even allowing for systematic underreporting, suicide is still rare under the age of 16 (Brooksbank, 1985).
Triolo, McKenry, Tishler, and Blyth (1984) noted that intrapersonal variables are good predictors of adolescent suicidal tendencies. Robbins and Alessi (1985) found relationships between self-esteem and suicidal behavior; Wetzel (1975, 1976) observed negative self-image in persons who had threatened or attempted suicide; and Wenz (1976) reported that attempters have low self-esteem. De Man, Leduc, and Labreche-Gauthier (1993) found that adolescents with low self-esteem or a less internal locus of control have higher suicidal ideation and are more likely to commit suicide. Suicidal youngsters tend to doubt themselves and their capacity to cope with adjustive demands (Curran, 1986), frequently face problems they regard as beyond their control (Orbach, 1986), and often feel helpless (Peck, 1983). …