The baseline multigroup second‐order CFA model also fit the CAT Study data well (CFI = 0.97 TLI = 0.97; RMSEA = 0.06). As above, constraining the first‐order factor loadings and thresholds and the second‐order factor loadings and intercepts across groups resulted in a significant deterioration in model fit (P < 0.05). However, partial measurement invariance was again obtained by freeing the loading for the anal sex item and the intercept for the first‐order CSA factor, and additionally by freeing the loading of the first‐order CSA factor. The fit indices again suggested the model with partial MI fit the data well (CFI = 0.955, TLI = 0.972, RMSEA = 0.089). Overall, the CFA results in CAT Study are also consistent with assessment of comparable abuse constructs in males and females. Factor loadings on the first‐ and second‐order factors are shown in Tables S2 and 1 respectively. For the first‐order factors, standardized factor loadings ranged from, 0.72–0.94 for CPA, 0.55–0.94 for CSA and 0.64–0.99 for PPA. Standardized Loadings on the second‐order CTF were 0.84 for CPA, 0.64 for PPA, and 0.58 for CSA in women and 0.29 in men.
