LAUSR

This latent class analysis study used a bias-adjusted three-step approach to empirically identify mutually exclusive clusters of teacher professional qualifications based on commonly studied indicators of teacher quality. We then examined the relationship between cluster membership and the mathematics gains of adolescents at risk for mathematics difficulties (MD), including students with disabilities and those without disabilities. We identified students at risk for MD as those performing at or below the 25th percentile on the state exam. We empirically identified eight qualitatively distinct and interpretable teacher qualification clusters. Based on value-added models, we found that teachers in Cluster 5 had lower average math learning gains than their peers in clusters with the most experienced and credentialed teachers. Cluster 5 included teachers who were novice and early-career, traditionally prepared math majors. We observed significant effects for the end or grade scores for the middle grades but not the algebra scores of high school students, suggesting teacher cluster membership effects varied by grade level. We discuss limitations and offer implications for research and policy.