Supplementary materials for: Developmental Relations Between Math Anxiety and Arithmetic Fluency in Elementary School Students Across Six Time Points

Math anxiety is widely recognized as a barrier to mathematical learning; however, its causal relationship with performance remains a topic of debate. The present study examined whether math anxiety predicts poor arithmetic fluency or vice versa by applying both a traditional Cross-Lagged Panel Model (CLPM) and a Random-Intercept Cross-Lagged Panel Model (RI-CLPM) to longitudinal data from 866 students in grades 4–6 across six measurement points. The CLPM revealed a reciprocal relationship between math anxiety and arithmetic fluency, suggesting that the two influence each other over time. In contrast, the RI-CLPM indicated a unidirectional effect, where math anxiety predicted subsequent declines in arithmetic fluency, supporting the debilitating hypothesis. Although both models fit the data, the RI-CLPM demonstrated superior fit indices. When stable between-person differences are accounted for, the model reveals that within-person increases in math anxiety predict subsequent declines in arithmetic fluency, supporting the debilitating hypothesis. These findings highlight the importance of model selection in developmental research and underscore the role of math anxiety as a critical target for early interventions aimed at improving mathematical outcomes.

Math anxiety is widely recognized as a barrier to mathematical learning; however, its causal relationship with performance remains a topic of debate. The present study examined whether math anxiety predicts poor arithmetic fluency or vice versa by applying both a traditional Cross-Lagged Panel Model (CLPM) and a Random-Intercept Cross-Lagged Panel Model (RI-CLPM) to longitudinal data from 866 students in grades 4–6 across six measurement points. The CLPM revealed a reciprocal relationship between math anxiety and arithmetic fluency, suggesting that the two influence each other over time. In contrast, the RI-CLPM indicated a unidirectional effect, where math anxiety predicted subsequent declines in arithmetic fluency, supporting the debilitating hypothesis. Although both models fit the data, the RI-CLPM demonstrated superior fit indices. When stable between-person differences are accounted for, the model reveals that within-person increases in math anxiety predict subsequent declines in arithmetic fluency, supporting the debilitating hypothesis. These findings highlight the importance of model selection in developmental research and underscore the role of math anxiety as a critical target for early interventions aimed at improving mathematical outcomes.