TABLE 2

Estimated Levels and Rates of Change of Boys’ and Girls’ Engagement in Sex-Related Online Behaviors, Derived From Unconditional Multigroup Univariate LGM Models

Online BehaviorRangeInterceptLinear Slope
Mean (SE)σ2 (SE)Mean (SE)σ2 (SE)
Boys
 SEIM use1–62.45 (.07)***1.88 (0.18)***0.32 (.03)***0.13 (0.04)**
 Sexual information seeking1–61.69 (.04)***0.45 (0.10)***0.01 (.02)0.09 (0.03)**
 Cybersex0–40.10 (.02)***0.05 (0.02)*0.03 (.01)**0.01 (0.01)
 SNS use0–62.42 (.06)***1.64 (0.15)***−0.05 (.03)0.18 (0.04)***
Girls
 SEIM use1–61.09 (.02)***0.10 (0.03)**0.05 (.01)***0.05 (0.01)***
 Sexual information seeking1–61.20 (.02)***0.11 (0.04)**0.04 (.01)**0.02 (0.01)*
 Cybersex0–40.03 (.01)***0.01 (0.01)0.01 (.00)0.00 (0.00)
 SNS use0–62.96 (.06)***1.26 (0.14)***−0.03 (.02)0.10 (0.03)**
  • Because a few participants had missing data on online behavior variables in all 4 waves, N models for boys = 592 to 593, N models for girls = 535. σ2 represents the variance around mean levels and rates of change. Growth factor means in italics differed significantly between boys and girls as indicated by the Wald test of parameter constraints. Models had adequate fit statistics: comparative fit index, ≥0.94; root mean square error of approximation, ≤0.06. *P < .05, **P < .01, ***P < .001.