= 36.51, SD = 11.43) consisting of 206 men, 274 women, nine non-binary individuals, one demisexual girl and one trans man, in a relationship, participated in an online study via Prolific. The GMSEX fitted a unidimensional factor model. However, significant standardized residuals showed covariance not explained by the main factor. IRT results showed that the GMSEX violated the assumption of local independence, and the scores did not discriminate well at high sexual satisfaction scores and are biased towards lower levels of sexual satisfaction. Additionally, the GMSEX contains redundant items that provide very similar information about sexual satisfaction. Results of the current study suggest that the GMSEX has limitations only discovered through IRT. While more research and validity evidence is needed to further examine the interpretability of GMSEX scores, the current findings suggest that the GMSEX might not be very useful to assess individuals with high sexual satisfaction.
Hay et al. (Tue,) studied this question.