Students in public education are expected to attend school on a regular basis to learn. This study applied multiple linear regression models and binary logistic regression models to analyze the effect of risk of early chronic absence pattern on later math achievement by using student level longitudinal data. Results showed that students with small absence pattern performed significantly better both in state math assessments and had higher probability to achieve assessment proficiencies than students with chronic absence pattern. Findings in this study contribute to educators, policy makers, and other researchers by raising awareness of students' attendance in early grade levels and helping them understand the correlation between chronic absence in early childhood and students' future mathematics achievement.