Life Science Research Immersion Program Improves STEM-Specific Skills and Science Attitudes among Precollege Students

Abstract

A predicted rapid growth in science, technology, engineering, and math (STEM) careers demands a vast and talented workforce, but students most commonly abandon STEM majors within the first 2 years of college. Performance in introductory courses, scientific literacy, and the ability to critically reason are main predictors of retention in STEM, highlighting the importance of precollege and early college experience. The Life Science Research Immersion Program (LSRIP) is a novel science education model that focuses on the development of scientific research skills, thus preparing students for introductory college courses and beyond. To evaluate the efficacy of the LSRIP, pre- and postprogram assessments and surveys were administered to three precollege student cohorts. Scientific reasoning assessment scores improved by 4.70% in Summer 2019 (P < 0.01), 9.44% in Fall 2019 (P < 0.05), and 0.97% in Winter 2020 cohorts, with two of five questions showing statistically significant improvement. Surveyed attitudes toward science improved in 62.9% of questions across all cohorts. These results suggest that research immersion experiences are an effective educational instrument for improving and promoting scientific reasoning and attitudes among precollege students. To better prepare students for success in STEM higher education and careers, we recommend implementing LSRIPs to complement traditional precollege science curricula.

Keywords: STEM education; biology; ecology; gene expression; molecular biology; neurobiology; physiology; plant biology; precollege; research immersion.

FIG 1

Score differences between post- and preprogram scientific reasoning assessments by individual question and student in SU19 (A), FA19 (B), and WI20 (C) programs. Per heatmap, each row represents a student, and each column represents a question. The heatmap colors represent score changes as percentages of the total possible points of each question, where orange indicates an increase in score and blue indicates a decrease. Considering each student’s responses to each question as a data point, the quarterly pie charts indicate percentages of positive (orange), negative (blue), or no score (gray) changes. For all three quarters, there were more increases than decreases in these scores.

FIG 2

(A and B) Comparing post- versus preprogram scientific reasoning assessment scores, students significantly improved on Q4 in SU19 (A) and Q2 in FA19 (B). (C) No questions in WI20 had a significant change in score. Points indicate preprogram mean scores, and arrowheads indicate postprogram mean scores, represented as percentages of a perfect score for that question. Arrowheads facing right indicate an increase, and arrowheads facing left indicate a decrease in mean score for a question, from pre- to postprogram. One-way ANOVA with Bonferroni correction for multiple comparisons was performed for all data set post- versus preprogram scores, and statistical significance was determined through multiplicity corrected values. *, P < 0.05; **, P < 0.01.

FIG 3

TOSRA post- versus preprogram mean score differences by question (A to G) and by category (H) (10 questions per category), pooled across all three quarters, SU19 to WI20. For panels A to G, individual categories show bars corresponding to each individual question. For panel H, bars correspond to each category, as mean score differences for all questions in each category. TOSRA category letter abbreviations are represented by panels (A to G). Questions were scored on a scale from 1 to 5, with 1 being the maximum negative response and 5 being the maximum positive response. Increase or positive value = positive attitude change; decrease or negative value = negative attitude change. *, P < 0.05; ***, P < 0.001.

FIG 4

Heatmaps of changes in students’ pre- to postprogram TOSRA response values for each question within the seven TOSRA categories. Questions were scored on a scale from 1 to 5, with 1 being the maximum negative response and 5 being the maximum positive response. A heatmap per category shows combined responses from students across all quarters (SU19, FA19, and WI20). Columns represent individual questions in the TOSRA and each row represents a student (N = 55), with the order of students consistent across all heatmaps. The cell bar on the top of each heatmap shows the combined total change per question for all students. The color scale at the top of each panel corresponds to score changes indicated by each cell bar, while color scale at the bottom right (+4 to −4) corresponds to individual student-by-question cells (orange as a positive pre-to-post change and blue as a negative pre-to-post change in student response to each TOSRA question). Missing data are marked with an x.