Validation of Brazilian Version of the Sexual Desire Inventory 2 (SDI-2)

ABSTRACT Objective: To traslate and validate of the Brazilian version of the SDI-2. Methods: This was a cross-sectional study. The cultural adaptation considered the stages of initial translation, synthesis of translations, evaluation by a committee of experts from different regions of Brazil, back-translation, and pre-test. The content validity and psychometric proprieties was assessed. Results: Ten specialists participated in the cultural adaptation of the SDI-2. The content validity showed a Content Validity Ratio (CVR) ≥ 0.75 (p = 0.05). A total of 674 subjects participated in the field study. The Exploratory Factorial Analysis (EFA) presented factor loads ≥ 0.445, and commonalities ≥ 0.40; and two dimensions represented 77% of the total variance explained. The Confirmatory Factorial Analysis CFA presented X2/df = 4.265; the Root Mean Square Error of Approximation RMSEA = 0.110; the Non-Normed Fit Index NNFI = 0.946; the Comparative Fit Index (CFI) = 0.963; the Goodness of Fit Index GFI = 0.986; and the Adjusted Goodness of Fit Index AGFI = 0.979 for a two-factor model. The coefficient values for the total SDI-2 score were 0.91 for Cronbach's alpha, 0.91 for McDonald's Omega, and 0.97 for the Greatest Lower Bound GLB coefficients. The invariance between sexes was 0.01 for the ΔCFI and ΔRMSEA, showing model stability for these two populations. Conclusion: The Brazilian version of the SDI-2 is self-report, valid, reliable and invariant across sex.