![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
Abstract
The purpose of the current study was to establish the replicability of the component structure of the Sexual Dependency Inventory (SDI), a broadband, multidimensional measure of problematic sexual behaviors and preoccupations associated with sexual addiction or other sexual disorders, such as the paraphilias. The internal structure of the SDI was previously evaluated in a sexual addiction clinical sample of 626 men and 85 women, but the replicability of the component structure was not evaluated. The present study examined the internal structure of SDI responses in a very large (N > 1,000) clinical sample, and evaluated the replicability of the structure in a second sample of similar size. The structure of the SDI derived was consistent with the previous version (SDI-3.0), was coherent and psychometrically sound, and replicated well in a second sample. Based upon these results, recommendations were made for revisions to establish the Sexual Dependency Inventory—4th Edition (SDI-4.0) Behavior and Preoccupation scales to better match the component structure identified and improved psychometric performance of the measure. Psychometric properties of scores from the revised scales were reported, including estimates of internal consistency reliability from both samples. Clinical implications and suggestions for future research are discussed.
Notes
1Four Frequency items with rotated pattern loadings less than.30, but.29 or greater were retained, one each on four different components, because they were congruent with the component content and scale item-total correlation was adequate or better.
2Two Power items with rotated pattern loadings less than.30, but.27 or greater were retained, one each on two different components, because they were congruent with the component content and scale item-total correlation was adequate or better.
3More than 75.0% of participants denied any amount of engaging in any of the behaviors represented in these components.