Grade disparity between sections at UIUC

A Data Driven Discovery by , , , , , , , and
Contact/Corresponding Author: Wade Fagen-Ulmschneider*
Updated for Fall 2019 registration on · First published:

Introduction

One of the most frustrating situations to find yourself in is a course where all of your friends are in the "easier section". For most of us, it feels like this happens all of the time. As part of a growing set of GPA visualizations, students in the Spring 2017 section of CS 205: Data Driven Discovery took a deep dive into the GPA dataset to find out whether the perception of disparity of grades between different sections is true or not.

Using GPA data from the most recent six semesters (Fall 2015 through Fall 2018, including summers/winters when available), we found the distribution of every section/instructor group within every course. For example, Calculus I (MATH 221) has been taught by seven different primary instructors recently. We found the following distributions:

Legend
(Highest Avg. Grades)(Lowest Avg. Grades)
Instructors with average grades significantly lower than the average grade for a course have increasing red hues.
Middle
50%-tile
Next
12.5%-tile
Next
7.5%-tile
Final 5%
(95%/5%)
The darkest shading shows the median grades in a course, with each ligher showing grades further from the median.
External Impact

In the preparation of this work, we found other factors — including the time of day of the lecture and if the course was in-person or online — also have a major contribution to the final grades in a course. We hope others will dive deeper into these factors in the future.

Find Your Course

Type the subject for any course at UIUC (eg: CS for Computer Science) to find the disparity of grades between different sections of a course:

Interested in Data Science? Try the new math gened: Data Science Discovery (STAT 107)
Co-taught and created by the professor of this visualization -- no prereqs!

Love it? Share it and check out other Data Driven Discoveries