This manuscript describes data collection, cleaning, and quality control procedures used in a large-scale, longitudinal, in-school study of executive function skills (EFs) and academic achievement in middle childhood, Project iLEAD (in-school Longitudinal Executive Function and Academic Achievement Database). Assessments were administered in real-world educational settings in a large sample of third through eighth grade students (ages 7 to 14; N = 1,280) over two years, with eight data collection timepoints in group settings. We assessed students with a novel, mobile EF assessment tool Adaptive Cognitive Evaluation (ACE). This battery included 11 tasks that were each designed to adapt to user performance in a trial-wise manner, allowing the same battery of tasks to be used multiple times within the same individual, and across a wide range of ages. Data quality analyses revealed that the adaptive algorithms were successful in equating challenge levels across ages 7 through 14 for 10 of 11 tasks. Data for each task were found to be approximately normally distributed and split-half reliability was acceptable across both accuracy and reaction time. ACE thus provides a reliable way to assess EFs in middle childhood using the same tasks while maintaining appropriate challenge level without facing ceiling or floor effects.