by Chris Gasser
Hi Everyone!
Your friendly Supplemental Instruction (SI) Coordinator here to tell you about some awesome findings in the SI world. First, just as a quick reminder, SI offers group study tables for challenging courses at OSU. Spring term registration is open, and we still have plenty of space for students to join remote tables. Now on to the exciting things.
During the past year, SI teamed up with Dr. Nicholas Martens from Institutional Research to answer some tough SI questions: Is there selection bias in SI? How much does SI actually help students? At what point should we consider a student an SI student? Using the past 4 years of data, including data from BI 21x, BI 23x, BI 33x, CH 23x, MTH 251, and PH 20x, we have answers for you.
In what ways, if any, are SI students different than Non-SI students?
Before we look at the impact of SI, many people ask about selection bias. Are students who elect to participate in SI different from students who don’t participate? Looking at the data, the following differences were found to be significant between the two populations. SI students are disproportionately female, have higher OSU cumulative GPAs and high school GPAs, have lower ALEKS math placement scores, and lower SAT/ACT scores. When it came to ethnic category, international status, first-generation status, age (as a binary of <25 & 25+), and Pell eligibility, there was no practically significant difference. These findings demonstrate that SI is serving many students proportionally across most demographics, while also reminding us that caution is needed when comparing SI to non-SI students.
Do students retaking courses and participating in SI earn higher grades?
One analysis looked at students who retook an SI supported course but who did not complete SI on their first course attempt. This analysis used a matched pair-design, matching by student and sorting into groups based on whether or not the student used SI on their second course attempt. While students typically do better on their second course attempt, this analysis showed that students retaking a class who used SI in their second attempt not only earned higher course grades their second time through, but earned over a half of a course grade higher than students who retook the course without using SI.
A second analysis conducted using linear regression produced a noteworthy inferential finding: the coefficient of impact on student course grade was .09 per SI attendance. In SI, we claim that students who complete SI earn on average 1/3 to ½ a grade point higher than non-SI students. The .09 coefficient x 4 times of SI attendance = .36 average course grade increase, falling right above that 1/3 course grade point and further supporting the claim that participation in SI increases average course grades.
These analyses provide strong evidence that SI really benefits students. Unsurprisingly, the more a student participates in SI, the greater the impact of the program. I’d love to share more of our findings with you, or talk about how we can get more students to experience the benefit of SI. Please don’t hesitate to reach out to me at chris.gasser@oregonstate.edu.
