Abstract Adult sex estimation is one of the first and most important steps in forensic examination. While dealing with disturbed burials, the most dimorphic anatomic areas of the skeleton (such as the coxae, the skull and the head of femur and humerus) may be deteriorated or fragmented. In contrast, the minimum supero-inferior femoral neck diameter (SID) is generally much better preserved. The aim of the present research is to identify the discriminatory potential of SID for sex estimation and to test different formulae and mathematical procedures currently available in the forensic literature, on a sample of 295 contemporary individuals from the 21st Century Identified Skeletal Collection (University of Coimbra, Portugal), in order to identify its relevance for application in Portuguese forensic cases. Results showed that SID is a dimorphic variable, with high frequencies and probabilities of cases correctly estimated (0.82 and 0.83, respectively); statistically significant differences between females and males, and a high association between the metrics and sex, were identified. Posterior probabilities allow reliable estimations for all the measurements, excepting those between 31.0 and 31.5 mm, and the procedures that show the highest accuracy are those proposed by Seidemann et al. (1998), Curate et al. (2016) and Luna et al. (2021), with frequencies and probabilities between 0.82 and 0.83 for both sexes. The validation procedures implemented in this study highlight both the need to test quantitative models generated from diverse contemporary human populations, and the value of SID for obtaining reliable adult sex estimates, as they improve the quality of the biological profiles obtained in forensic contexts. Key points