Utilizing a cross-sectional sample of young Chinese children in Guangdong (Study 1; N = 609) and a longitudinal sample of young Chinese children in Hong Kong (Study 2; N = 197), two studies were conducted to validate the Preschool Affective Attitudes toward Mathematics Scale (PAAMS). In each study, children were interviewed individually on their affective attitudes toward mathematics activities with their parents, teachers, and classmates and were tested individually on their mathematics performance. Parents also rated children's interest in mathematics activities. Both studies showed that a three-factor model that captured young children's affective attitudes toward mathematics activities with their parents, teachers, and classmates fitted the data best. The PAAMS demonstrated acceptable internal consistency reliabilities and satisfactory convergent correlations with parent-report child mathematics interest. Child-report mathematics affective attitudes were positively associated with mathematics performance. Moreover, Study 2 provided further evidence for the stability of the PAAMS's reliabilities and convergent and criterion validity and strict measurement invariance across the two-year follow-up period. The results suggest that the PAAMS is a reliable, valid, and developmentally suitable instrument.