The present study examined the factorial invariance and psychometric properties of the Snijders–Oomen Nonverbal Intelligence Test (SON-R 2.5-7) among boys and girls aged 2.5 to 7 years. Establishing equivalence across demographic groups, particularly gender, is essential for valid interpretation and comparison of test scores. The target population included all children aged 2.5 to 7 years residing in Tehran in 2005, and the final sample consisted of 1,126 children (615 boys and 511 girls), selected through multistage random sampling from kindergartens affiliated with the Tehran Welfare Organization and the Ministry of Education. Data were analyzed using R statistical software with advanced psychometric and multivariate techniques. Exploratory factor analysis was first conducted to identify the underlying structure of the test, and findings supported a two-factor model representing performance and reasoning abilities. This structure was then evaluated through confirmatory factor analysis, which showed acceptable goodness-of-fit indices and confirmed the adequacy of the proposed model. To determine whether the measurement model functioned equivalently across gender groups, multigroup confirmatory factor analysis (MGCFA) was used. Measurement invariance was assessed at successive levels, including configural, metric, scalar, and strict invariance. The results demonstrated that the SON-R 2.5-7 met the required criteria for invariance between boys and girls, indicating that the test operates similarly across gender. Because measurement invariance is a central source of evidence for construct validity, these findings provide strong support for the construct validity of the instrument. Overall, the SON-R 2.5-7 can be considered a valid, reliable, and gender-unbiased tool for assessing and comparing nonverbal intelligence in children aged 2.5 to 7 years in both research and clinical settings.