Ian Montgomery

The Ohio Education Research Center — a collaborative of the John Glenn College and the Center for Human Resource Research — offers the Public Sector Data Internship Program for students interested in exploring careers in policy research and evaluation. Students receive instruction on data-based methods to investigate policy problems, and work at a state agency in Columbus for the summer. The program allows students to discover the exciting work that the Ohio Education Research Center and its affiliate partners are engaged in and gives students practical experience with theory they learn in college classes. This summer we have 5 interns participating in the program. To highlight their hard work and dedication this summer, the OERC will be featuring each of our summer interns.

Our next intern of the week is Ian Montgomery. Mr. Montgomery is a third year Economics Major from Columbus, Ohio. He is interning with the Ohio Department of Education to on an analysis to identify positive outlier schools in Ohio. His reflections on the internship are shared below:

My name is Ian Montgomery, and I will be going into my third year at OSU this upcoming fall. I am majoring in Economics with a minor in English. This summer, I have been working with the Ohio Department of Education on a project identifying positive outlier schools. The main goal of this project is to identify schools across the state of Ohio that are performing extraordinarily well.

Each week, I work at the John Glenn College of Public Affairs building with the OERC’s staff. I typically meet with the OERC’s team two to three times per week to create code and visualizations in Tableau, R, and Stata. I have been using district and school level data collected through the Education Management Information System (EMIS) to determine which variables in the datasets I work with are highly correlated with each other. EMIS is a statewide system used to collect data from Ohio’s primary and secondary education system.

My positive outliers project relies heavily on linear regression and predicted models that I have created. These models utilize Math and English proficiency rates as dependent variables to measure the expected performance of each district and school. The independent variables in my models include the rates of: students enrolled in Medicaid, students with disabilities, students who are English learners, and minority rates. Each model produces an actual and predicted value that is then subtracted from each other and put into a ranking system. There are over 600 districts in Ohio and nearly 4,000 schools, so working with large datasets requires a lot of patience and analytical skills. Figuring out how to use each dataset effectively has been a very informative experience for me throughout this internship.

My goal during this program was to become more familiar with data science as a whole. My background is in economics, but I feel like I have accomplished so much as the internship is wrapping up. The best part of this program was being able to meet with guest speakers each week and hear their experiences in data-oriented careers. This experience has been an extremely valuable learning opportunity, and I would highly recommend participating in this program for anyone interested in the public sector or data science.