Each iteration of high stakes accountability has included requirements to include measures of attendance in their accountability programs, thereby increasing the salience of this measure. Researchers too have turned to attendance and chronic absence as important outcomes in evaluations and policy studies. Often, too little attention is paid to the implications of measurement or statistical modeling decisions employed in these studies. Such lack of attention to measurement and modeling can obscure important differences in these measures across the educational life cycle, as well as between the outcomes of higher income students and their lower-income peers. Using longitudinal data from a representative state, this paper demonstrates how measurement and modeling choices can influence model estimates and undermine the quality of inferences. These considerations have important implications for researchers and practitioners who wish to use attendance data to understand policy impacts and guide practice, particularly when focused on chronic absence.