Bounds on Teacher Survey Outcomes
| Original Estimates | Upper Bounds | Lower Bounds | N (Original) | N (Lee Bounds) | ||||
|---|---|---|---|---|---|---|---|---|
| General Text Treatment (1) | Personalized Text Treatment (2) | General Text Treatment (3) | Personalized Text Treatment (4) | General Text Treatment (5) | Personalized Text Treatment (6) | |||
| Teacher report of interactions factor | 0.2162 (0.1333) | −0.1739 (0.1537) | 0.3313* (0.1264) | −0.0018 (0.1650) | 0.0036 (0.1315) | −0.1687 (0.1909) | 368 | 339 |
| Parent talks about child’s interests | 0.2312+ (0.1185) | −0.0892 (0.1528) | 0.3666* (0.1765) | 0.0678 (0.1866) | 0.0584 (0.0839) | −0.2497 (0.1617) | 396 | 367 |
| Parent asks how well child gets along with others | 0.1982 (0.1580) | −0.03 (0.1727) | 0.3729* (0.1730) | 0.1825 (0.1943) | 0.0859 (0.1663) | −0.2598 (0.1724) | 395 | 367 |
| Parent asks how well child is doing in school | 0.1982 (0.1585) | −0.1357 (0.1901) | 0.2412 (0.1629) | 0.0965 (0.1907) | 0.0326 (0.1765) | −0.2593 (0.1866) | 397 | 367 |
| Parent asks about child’s literacy skills | 0.2553+ (0.1321) | −0.1751 (0.1128) | 0.3531* (0.1395) | −0.0255 (0.1159) | −0.0282 (0.1391) | −0.3225** (0.1060) | 396 | 367 |
| Parent asks how to help child learn to read | 0.2063 (0.1490) | −0.1775 (0.1579) | 0.3770* (0.1594) | 0.0295 (0.1728) | −0.0992 (0.1303) | −0.4491** (0.1310) | 397 | 367 |
| Parent asks for book recommendations | 0.0251 (0.1273) | −0.1528 (0.1438) | 0.0993 (0.1538) | −0.0574 (0.1705) | −0.1815 (0.1316) | −0.3850** (0.1371) | 396 | 367 |
| Parent talks about home activities | 0.2787* (0.1131) | −0.0571 (0.1720) | 0.3770** (0.1182) | 0.1502 (0.1535) | 0.1132 (0.1067) | −0.2381 (0.1882) | 409 | 379 |
| How well does teacher know parent | −0.0762 (0.1709) | −0.2610+ (0.1527) | 0.0959 (0.1671) | −0.0827 (0.1851) | −0.2474 (0.1805) | −0.3581* (0.1659) | 382 | 351 |
Notes: Each pair of cells represents the results of a separate regression of the treatment effect on the relevant outcome. Row headers indicate the outcome. Column headers indicate model components. Column 1 and 2 present the original estimates from Columns 3 and 4 of Panel C of Table 5. All outcomes are standardized. Factor analysis was used to create the outcome factors. Upper and lower bound estimates were calculated by the procedure recommended by Lee (2009). 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 in Panels A and C are the responses to teacher surveys fielded in May of 2015. Standard errors are clustered at the randomization site level.
↵+ p < 0.10,
↵* p < 0.05,
↵** p < 0.01, ***p < 0.001.