"Double-Dose" Algebra as an Alternative Strategy to Remediation: Effects on Students' Academic Outcomes
Nomi T, Allensworth E · 2009
grade Cquasi-experimentindependentmixed
Sample
Chicago Public Schools 9th-grade cohorts around the 2003 double-dose cutoff
Population
CPS 9th graders scoring below the national median on 8th-grade math, who were assigned a second period of algebra support.
Design
Interrupted time series plus comparisons around the 50th-percentile assignment cutoff for the 2003 policy. Short-run outcomes only; the credible long-run estimates come from the later regression-discontinuity work.
Key findings
Adding TIME AND SUPPORT at the right level — a second algebra period instead of remediation — moved achievement where the 1997 mandate had not. The gain is smallest for the weakest students, which is the readiness-floor caveat.
Genetic confound
Assignment is on measured prior achievement, so the treated group differs from the untreated by construction; the cutoff comparison is what limits the damage.
Replication notes
Short-run pessimism about this policy was superseded by the long-run regression-discontinuity findings (Cortes, Goodman & Nomi), which found large attainment gains.
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Algebra achievement, first year of the policy | outcome | test scores improved; share earning B or better +9.4pp | standardized | end of 9th grade | active-alternative | end-of-treatment | domain-skill |
| Who benefited | pattern | very-low-skill students benefited least — a readiness floor, not a universal lever | standardized | end of 9th grade | active-alternative | end-of-treatment | domain-skill |
Cited by
- Algebra timing — acceleration mandates, delay mandates, and readiness-based placementmixedconf: highgc: low