Investigating the impact of peer supplemental instruction on underprepared and historically underserved students in introductory STEM courses

Abstract

Background: Supplemental instruction (SI) is a well-established mode of direct academic support, used in a wide variety of courses. Some reports have indicated that SI and similar peer-led academic support models particularly benefit students identifying with historically underserved racial/ethnic groups in STEM. However, these studies have not explicitly examined the role of prior academic experiences, an important consideration in college success. We report on the impact of a modified SI model, Peer Supplemental Instruction (PSI), on student success in introductory STEM courses at a diverse access institution. This study focuses on PSI's impact on the academic performance of students identifying with historically underserved racial/ethnic groups, while also considering the effects of prior academic experiences.

Results: Data were aggregated for nine courses over five semesters to produce a robust data set (n = 1789). PSI attendees were representative of the overall student population in terms of previous academic experiences/performance (as determined by high school GPA) and self-identified racial/ethnic demographics. Frequent PSI attendance was correlated with a significant increase in AB rates (average increase of 29.0 percentage points) and reduction in DFW rates (average decrease of 26.1 percentage points) when comparing students who attended 10 + vs. 1-2 PSI sessions. Overall, students identifying as Black/African American received the largest benefit from PSI. These students experienced a significant increase in their final course GPA when attending as few as 3-5 PSI sessions, and exhibited the largest increase in AB rates (from 28.7 to 60.5%) and decrease in DFW rates (from 47.1 to 14.8%) when comparing students who attended 10 + vs. 1-2 sessions. However, students with similar HS GPAs experienced similar benefits from PSI, regardless of self-identified race/ethnicity.

Conclusions: The data presented here suggest that PSI particularly benefitted underprepared students in their introductory STEM courses. Since students identifying with historically underserved racial/ethnic groups have traditionally had inequitable K-12 educational experiences, they enter college less prepared on average, and thus particularly benefit from PSI. PSI, in conjunction with additional strategies, may be a useful tool to help rectify the results of systemic educational inequities for students identifying with historically underserved racial/ethnic groups.

Keywords: Academic preparedness; Diversity; Equity; Gateway courses; Historically underserved; STEM education; Student success; Supplemental instruction.

Fig. 1

Self-identified racial and ethnic demographic distributions of all GGC students (N = 12,831) compared to students enrolled in STEM majors at GGC (N = 3640) in Fall 2019

Fig. 2

a Self-identified racial/ethnic demographic distribution of students enrolled in PSI-supported courses vs. students attending PSI sessions. b Self-identified racial/ethnic demographic distribution of students attending PSI sessions, separated by frequency of attendance (1–2, 3–5, 6–9, or 10 + PSI sessions for a given course within a single semester)

Fig. 3

Final course grade distributions of PSI participants, grouped by session attendance, for students identifying as a Asian (N = 265), b Black/African American (N = 679), c Hispanic/Latino (N = 401), and d White (N = 444). Shown for comparison are the final grade distributions for all students enrolled in PSI courses for students identifying as a Asian (N = 2834), b Black/African American (N = 6238), c Hispanic/Latino (N = 4143), and d White (N = 5375)

Fig. 4

Aggregated a AB and b DFW rates for PSI attendees as a function of PSI attendance and students’ identified race/ethnicity

Fig. 5

Mean final course GPAs of PSI participants (N = 1789) as a function of PSI attendance, and of all students enrolled in PSI-supported courses (N = 18,590), for students identifying with the four racial/ethnic groups studied here. Points are aggregate means ± SEM of student final course GPAs from nine foundation courses over 5 semesters

Fig. 6

Comparison of the effects of PSI participation on final course GPA for students identifying with the four racial/ethnics group investigated here, separated by incoming HS GPA: a < 2.5 (n = 307), b 2.5–3.5 (n = 829), and c > 3.5 (n = 318)

Fig. 7

Self-identified racial/ethnic demographic distribution of students attending PSI sessions, separated by HS GPA, and compared to the self-identified racial/ethnic demographic distribution of all PSI attendees