This repository contains web guides, R code, and sample data to help education data analysts learn how to apply predictive analytics to education data analyses. This material can also be accessed on the OpenSDP website at opensdp.github.io/analysis.
This guide provides an overview of common tools and techniques for implementing predictive analytics on education data. Synthetic student-level data is provided to illustrate how to go from exploring longitudinal student attainment data to developing and evaluating multiple machine learning models of student high school completion.
After completing this guide, the user will be familiar with the most common principles and technique of predictive analytics and how they can be applied to education data analysis problems - particularly through the example of student "early warning indicators".
This guide uses synthetic data created by the OpenSDP synthetic data engine. The data reflects student-level attainment data and is organized to be similar to the level of detail available at a state education agency - with a single row representing a student-grade-year. This guide does not cover the steps needed to clean raw student data. If you are interested in how to assemble data like this or the procedure used to generate the data, the code used is included in the data subdirectory.
This repository is organized in the following folder structure:
code
contains the R code of the analysis (and renders the guide)R
contains R functions necessary to complete the analysisdata
contains data necessary to complete the analysisimg
contains images to format the guidesincludes
contains style assets to format the guidesman
contains the manuals and descriptive files of the data
These materials were originally authored by Jared Knowles on November 21, 2018.
OpenSDP is an online, public repository of analytic code, tools, and training intended to foster collaboration among education analysts and researchers in order to accelerate the improvement of our school systems. The community is hosted by the Strategic Data Project, an initiative of the Center for Education Policy Research at Harvard University. We welcome contributions and feedback.