Quantile regression revealed larger intervention effects on GMRT reading comprehension at the .25–.30 quantiles (g = .35, ~9 points) and the .50 quantile (g = 0.22, ~5 points), but not at other quantiles or on WJ-III Passage Comprehension¶

Claim · All claims
Evidence · 1 study (3 entries) · causal r2 · q3 · i1 small
Currency · newest study 2018 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i1 At the .50 quantile of GMRT reading comprehension, the treatment group outperformed the comparison group by approximately 5 points (g = 0.22), controlling for baseline. → Al Otaiba 2018 q3 i1 At the .25 to .30 quantiles of GMRT reading comprehension, the treatment group outperformed the comparison group by approximately 9 points (g = .35). → Al Otaiba 2018 (2) q3 i? No significant differences were observed at other estimated quantiles of the GMRT, nor on the WJ-III Passage Comprehension outcomes at any quantile. → Al Otaiba 2018 (3)

Evidence¶

Al Otaiba 2018¶

Al Otaiba, S., Petscher, Y., Wanzek, J., Lan, P., & Rivas, B. (2018). I'm not throwing away my shot: What Alexander Hamilton can tell us about standard reading interventions. Learning Disabilities Research & Practice, 33(3), 156–167. https://doi.org/10.1111/ldrp.12179

q3 · i1 · causal · r2

Exploratory post-hoc multilevel quantile regression on GMRT reading comprehension post-test, estimated with the lqmm package in R with student-within-school nesting. The article reports "an effect size of g = 0.22" at the .50 quantile.

"Specifically, at the .50 quantile, there was an effect size of g = 0.22, and the treatment group outperformed the comparison group by approximately 5 points, controlling for baseline."

Al Otaiba 2018 (2)¶

Al Otaiba, S., Petscher, Y., Wanzek, J., Lan, P., & Rivas, B. (2018). I'm not throwing away my shot: What Alexander Hamilton can tell us about standard reading interventions. Learning Disabilities Research & Practice, 33(3), 156–167. https://doi.org/10.1111/ldrp.12179

q3 · i1 · causal · r2

Same exploratory quantile regression analysis. The article reports "a larger effect of Passport" at the .25 to .30 quantiles, with the treatment group ahead by ~9 points (g = .35).

"At lower levels of the GMRT Reading post-test (e.g., .25 to .30 quantiles), there was a larger effect of Passport, where the treatment group outperformed the comparison group by ~9 points (g = .35)."

Al Otaiba 2018 (3)¶

Al Otaiba, S., Petscher, Y., Wanzek, J., Lan, P., & Rivas, B. (2018). I'm not throwing away my shot: What Alexander Hamilton can tell us about standard reading interventions. Learning Disabilities Research & Practice, 33(3), 156–167. https://doi.org/10.1111/ldrp.12179

q3 · i? · causal · r2

Same quantile regression analysis of both comprehension measures. The article reports no significant differences at other GMRT quantiles and none at any quantile on WJ-III Passage Comprehension; equivalence was not tested.

"No significant differences were observed at other estimated quantiles of the GMRT, nor on the WJ-III PC outcomes at any quantile."

Discussion¶

Related Claims¶

• Students' relative rank order in reading remained stable across fourth grade on most measures, with GMRT reading comprehension the exception (r = .16) — related
• Effect sizes for the intervention were less than g = .03 on most outcomes, with small comprehension effects of g = 0.15 and g = 0.19 — related
• GAMLSS is preferred over quantile regression when the total outcome distribution is of interest, while quantile regression suits specific quantile regions — related
• SKILL does not improve standardized reading comprehension overall, with a small low-average effect at follow-up — related
• Passport to Literacy produced a small positive but nonsignificant effect on GMRT reading comprehension (g = 0.28) relative to business-as-usual comparison — related
• Quantile regression showed significant Passport to Literacy effects only between the .40 and .70 quantiles of GMRT posttest comprehension — related
• Quantile regression estimates show little evidence of heterogeneous Mathematics Mastery treatment effects across the post-test score distribution — related