The baseline multigroup second‐order CFA fit the CT Study data well (CFI = 0.976, TLI = 0.974, RMSEA = 0.029). Constraining the first‐order factor loadings and thresholds and the higher order factor loadings and intercepts across gender indicated that measurement invariance (MI) did not hold (P < 0.001). However, partial MI was obtained by freeing the loading of a single item assessing attempted or completed anal sex, and freeing the intercept for the first‐order CSA factor (P = 0.091). In addition, fit indices suggested the model with partial MI also fit the data well (CFI = 0.981, TLI = 0.981, RMSEA = 0.025). For the first‐order factors, standardized factor loadings (see Table S2) ranged from, 0.69–0.94 for CPA, 0.58–0.97 for CSA and 0.69–0.96 for PPA. Standardized Loadings on the second‐order factor (CTF) were 0.82 for CPA, 0.69 for PPA, and 0.50 for CSA (Table 1).
