Suicidal ideation and self-injury in LGB youth: a longitudinal study from urban Switzerland

Table 2 Frequencies of suicidal ideation (SUI) and self-injury (SI) (ever vs. never in the past month) for female and male participants at the three measurement waves (M = 15.4 years of age, M = 17.4 years of age and M = 20.6 years of age)

Ratings were provided on a 5-point Likert Scale ranging from never (1) to very often (5). Ratings were dichotomized for analyses (ever = 2–5 / never = 1). The past month was used as reference for the ratings
Note 2. Statistical parameters of the 2 (gender: female vs. male) × 3 (time-point: 15 years vs. 17 years vs. 20 yrs) mixed ANOVAs for SUI and SI:
SUI: sig. main effect of gender, F(1, 1087) = 16.62, p < .001, η²p = .02, with females (M = .25, SD = .43) reporting suicidal ideations more often than males (M = .17, SD = .38). Non-significant main effect of time-point, F(2, 2174) = 2.63, p = .07, η²p = .01. Significant interaction between gender and time-point, F(2, 2174) = 9.68, p < .001, η²p = .01. The follow-up pairwise comparisons revealed no significant effects, mean difference ≤ .01, p_s ≥ .06
SI: sig. main effect of gender, F(1, 1090) = 227.18, p < .001, η²p = .17, with females (M = .14, SD = .34) reporting self-injury more often than males (M = .06, SD = .23). Significant main effect of time-point, F(2, 2180) = 4.29, p < .05, η²p = .01, showing that self-injury at time-point 1 (15 years; M = .11, SD = .32) was significantly higher than at time-point 3 (20 years; M = .08, SD = .27). No significant interaction between gender and time-point resulted, F(2, 2180) = 1.69, p = .19, η²p = .01

ISSN: 1753-2000