Results from the Hypothesis Test that the Disruption in each Observed Covariate is Zero in the Population at each of the Three Cut-Scores, from a Seemingly Unrelated Regression (SUR) Regression-Discontinuity Model where each Covariate is Treated as an Outcome, for the Full Sample of Urban, Low-Income Students and the Subsample of Students who Plan to Attend a Four-Year College
| Urban, Low-Income Students | Urban, Low-Income Students Without Four-Year College Plans | |
|---|---|---|
| Panel I: Grade 8 | ||
| Needs Improvement/Warning cutoff | χ2(11) = 11.15 p = 0.431 | χ2(11) = 6.63 p = 0.829 |
| Proficient/Needs Improvement cutoff | χ2(11) = 10.28 p = 0.505 | χ2(11) = 9.02 p = 0.620 |
| Advanced/Proficient cutoff | χ2(11) = 6.55 p = 0.835 | χ2(11) = 10.60 p = 0.477 |
| Panel II: Grade 10 | ||
| Proficient/Needs Improvement cutoff | χ2(11) = 7.59 p = 0.749 | χ2(11) = 6.90 p = 0.807 |
| Advanced/Proficient cutoff | χ2(11) = 15.66 p = 0.154 | χ2(11) = 9.96 p = 0.534 |
Notes: All inferences from two-tailed hypothesis tests. Covariates treated as outcomes include student race, gender, whether the student was new to the state, was currently or formerly classified as limited English proficient, or required special education.