Table 2.
Linear Probability Models (LPM) predicting who takes calculus and who fails calculus.
| Taken Calculus | Failed Calculus | |
|---|---|---|
|
|
||
| Compared to Students Who Never Took Calculus |
Only among Students Who Took Calculus |
|
| Demographics | ||
| Female | −0.11 *** (−8.20) | −0.02 (−0.44) |
| Age | −0.38 (0.11) | −0.76 (−1.46) |
| Age squared | 0.01 (0.11) | 0.02 (1.50) |
| Black | 0.01 (0.68) | −0.01 (−0.81) |
| Hispanic | 0.01 (0.60) | 0.01 (0.15) |
| Asian | 0.09 * (2.31) | 0.07 (0.84) |
| Socio-economic status composite | 0.02 * (2.21) | −0.06 * (−2.11) |
| Prior academic skills and achievement | ||
| NELS 12th grade test score percentile (logged) | 0.04 *** (4.69) | −0.03 (−0.45) |
| High school GPA (logged) | 0.11 *** (4.27) | −0.17 + (−1.70) |
| Highest math course taken in High School | ||
| Geometry | −0.03 (−1.64) | −0.20 (−1.06) |
| Algebra II | −0.02 + (−1.76) | 0.07 (−0.35) |
| Trigonometry | 0.04 + (1.90) | −0.07 (−0.37) |
| Pre-calculus | 0.10 *** (3.77) | −0.11 (−0.59) |
| Calculus | 0.31 *** (8.75) | −0.12 (−0.65) |
| Planned to major in STEM | 0.13 *** (7.23) | 0.03 (0.74) |
| Institution Type | ||
| Private not-for-profit 4-year | 0.06 ** (2.62) | 0.05 (1.23) |
| Public 4-year | 0.03 * (2.18) | 0.11 * (2.37) |
| Constant | 3.37 (7.23) | 7.53 (1.55) |
| R2 | 0.24 | 0.11 |
| n | 3490 | 540 |
Source: National Educational Longitudinal Study (NELS:88), Postsecondary Education Transcript Study (PETS:2000) (NCES 1988; NCES 2000). t-statistics underneath coefficients in parentheses. Controls are in reference to male, White, highest math course taken as Algebra I or other math course in high school, and entered a public two-year college. Sampling weight used in analyses. n in models have been rounded to the nearest 10 for disclosure.
+
p < 0.1,
*
p < 0.05,
**
p < 0.01,
***
p < 0.001.