Quantile regression showed significant Passport to Literacy effects only between the .40 and .70 quantiles of GMRT posttest comprehension¶

Claim · All claims
Evidence · 1 study · causal r3 · q2 · i1 small
Currency · newest study 2016 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i1 Significant treatment effects on GMRT reading comprehension occurred from approximately the .40 to the .70 quantile, with effect sizes ranging from g = 0.23 to g = 0.38. → Wanzek et al. 2016

Evidence¶

Wanzek et al. 2016¶

Jeanne Wanzek, Yaacov Petscher, Stephanie Al Otaiba, Shawn C. Kent, Christopher Schatschneider, Martha Haynes, Brenna K. Rivas, Francesca G. Jones. (2016). Examining the Average and Local Effects of a Standardized Treatment for Fourth Graders with Reading Difficulties. https://eric.ed.gov/?id=ED577133

q2 · i1 · causal · r3

Exploratory quantile regression of GMRT posttest on the treatment dummy code, reported in Figure 1b. At the .40 quantile the intervention coefficient was 8.59 (t(1) = 8.59, p < .05), corresponding to g = 0.28; averaged across the .40 to .60 quantile range the mean effect size was g = 0.32 (SD = 0.05).

"The panel in Figure 1b illustrates that significant effects for the Passport to Literacy intervention were observed from approximately the .40 quantile to the .70 quantile. Effect sizes within this part of the distribution of the GMRT reading comprehension ranged from g = 0.23 at the .60 quantile up to g = 0.38 at the .50 quantile."

Discussion¶

Related Claims¶

• 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 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 — related
• GAMLSS is preferred over quantile regression when the total outcome distribution is of interest, while quantile regression suits specific quantile regions — related
• Chatbot treatment effects on final grades were statistically significant between the 35th and 75th quantiles of the course grade distribution and at the 95th — related
• Quantile regression estimates show little evidence of heterogeneous Mathematics Mastery treatment effects across the post-test score distribution — related