This study focuses on the psychometric properties of the Romanian adaptation of the Sexual Risk Survey (SRS), a self-reported measure of risky sexual practices. Using a large sample, this study investigated various forms of reliability and validity of the SRS. Results indicated good internal consistency coefficients and a good convergent and discriminant validity for SRS. The confirmatory factor analysis displayed acceptable fit indices for the five-factor model. Results obtained via Item Response Theory showed that SRS scales offer good information for trait level only for one of the five scales and poor coverage for trait levels lower than 0 for the remaining scales. Network analysis showed quite similar centrality indices for all SRS items, both for men and women. However, there are some differences between genders in terms of the type of association between several items. Thus, the current study turned light on the SRS from an IRT and network analysis perspective.