Heterogeneity of Academic Outcomes by Baseline Academic Performance (First Year of Experiment Only)
| General Text Treatment (1) | Personalized Text Treatment (2) | |
|---|---|---|
| Panel A: Quartiles 2 and 3 (N = 123) | ||
| Reading level (ordinal logit) | 0.3545 (0.3405) | 0.4253 (0.3780) |
| Reading level (standardized point scale) | 0.0645 (0.1095) | 0.0895 (0.0817) |
| Exceeds expectations | 0.0092 (0.1089) | 0.0495 (0.0799) |
| Meets or exceeds expectations | 0.1713+ (0.0882) | 0.2196 (0.1346) |
| Approaches, meets, or exceeds expectations | 0.051 (0.1015) | 0.046 (0.1515) |
| Panel B: Quartiles 1 and 4 (N = 121) | ||
| Reading level (ordinal logit) | −0.3068 (0.7124) | 0.9128* (0.4183) |
| Reading level (standardized point scale) | −0.1222 (0.2898) | 0.3631 (0.2348) |
| Exceeds expectations | −0.0722 (0.1048) | 0.2154* (0.0844) |
| Meets or exceeds expectations | −0.0934 (0.1169) | 0.0728 (0.0824) |
| Approaches, meets, or exceeds expectations | −0.0559 (0.1258) | 0.0361 (0.0795) |
Notes: Each pair of cells represents the results of a separate regression of the treatment effect on the relevant academic outcome. Column headers indicate the model components. The reference category is the control group. Row headers indicate the academic outcome. Panel headers indicate the quartile subsample. All models include randomization site fixed effects, controls for texting language, factors of baseline survey responses, and administrative covariates. Covariates are detailed in Table 3. Source data are district test files of the Fountas and Pinnell Benchmark Assessment System in fall of first grade. Standard errors are clustered at the randomization site level.
↵+ p < 0.10,
↵* p < 0.05, **p < 0.01, ***p < 0.001.