The Sport Motivation Scale (SMS) is a widely used instrument that assesses motivational processes within sport and exercise environments. The scale has demonstrated validity and reliability in multiple cultures, however, there is yet no empirical evidence regarding its psychometric properties in the Czech population. In this study we therefore set out to examine the reliability and construct validity of the SMS in a sample of Czech university-aged athletes. We first examined the SMS factor structure using a nonparametric item response theory model (Mokken monotone homogeneity model) and identified six items violating the unidimensionality of the particular subscales. Remaining items were then subjected to test of hypothesized seven-factor structure and several different forms of measurement invariance examined based on gender, competition level and type of sport (individual vs team sports). The hypothesized seven-factor fit well and there was sufficient evidence supporting full invariance across the examined groups. All SMS subscales had adequate internal consistencies ranging from 0.66 to 0.89. Results of correlational analysis among the SMS subscales and between the SMS and two outcomes of interest further supported validity of the scale. Observed differences in SMS subscales between males and females, recreational and competitive athletes, as well as between individual and team-based sport activities, comported with prior empirical studies using a self-determination theory framework. In conclusion, results reinforce the utility and performance of the SMS in a sample of Czech university athletes. The SMS may therefore be recommended for measurement of the multidimensional motivational processes taking place in the exercise and sport domain.