Table 7 Logistic regression coefficients with dependent variable STEM retention after two years. Gateway A/B grades are the only significant predictor of STEM retention, and its effect size does not significantly differ pre- to post-implementation
From: Required peer-cooperative learning improves retention of STEM majors
Factor | Coefficient B | Significance p | Odds Ratio | 95% CI |
|---|---|---|---|---|
Pre-implementation Nagelkerke r 2 = 0.219 | ||||
Demographics | 0.306 | 0.132 | 1.36 | 0.91–2.02 |
SAT-Math (std.) | 0.702 | 0.054 | 2.02 | 0.99–4.12 |
SAT-Verbal (std.) | −0.399 | 0.143 | 0.671 | 0.39–1.14 |
Gateway A/B grade | 1.390 | <0.001 | 4.02 | 1.98–8.16 |
(Constant) | −0.894 | 0.001 | ||
Post-implementation Nagelkerke r 2 = 0.286 | ||||
Demographics | −0.218 | 0.163 | 0.804 | 0.59–1.09 |
SAT-Math (std.) | 0.157 | 0.526 | 1.17 | 0.72–1.90 |
SAT-Verbal (std.) | −0.015 | 0.943 | 0.986 | 0.66–1.47 |
Gateway A/B grade | 2.071 | <0.001 | 7.935 | 4.27–14.8 |
(Constant) | −0.0962 | <0.001 | ||