Walkthrough 3: Using school-level aggregate data to illuminate educational inequities

This chapter explores cleaning, tidying, joining, and visualizing publicly available aggregate data. Data scientists in education frequently work with public aggregate data when student-level data is not available. By working with these data, data scientists in education can discover broader trends and underlying patterns. If aggregate data is disaggregated by subgroups or subpopulations, data scientists can reveal areas of inequity for marginalized populations. Using a freely available district dataset, this chapter looks at the distribution of students in the district by race and socioeconomic status. Subgroup analysis can point out the state of equity in a system to inform how to improve the situation for more equitable opportunities for students. Data science techniques in this chapter include reading tables from an online PDF into a machine-readable format, preparing data for analysis, transforming it into a tidy format, visualizing it, and analyzing distributions and relationships.