Randomization Checks: The Effect of Treatment Status on Pretreatment Child Covariates (Pooled Sample)
| Pretreatment Child Covariates | Parent Survey Sample | Teacher Survey Sample | Academic Sample |
|---|---|---|---|
| Child female | 0.029 (0.043) | −0.05 (0.063) | −0.003 (0.035) |
| Hispanic | −0.011 (0.035) | −0.059 (0.045) | −0.025 (0.024) |
| Chinese | 0.013 (0.025) | 0.034 (0.033) | −0.002 (0.020) |
| Black | 0.017 (0.024) | 0.031 (0.044) | 0.033+ (0.019) |
| White | −0.037 (0.026) | −0.015 (0.027) | −0.013 (0.016) |
| Other race | 0.011 (0.025) | 0.009 (0.032) | 0.008 (0.021) |
| Child age in years | 0.023 (0.028) | 0.02 (0.042) | 0.052 (0.037) |
| Parent female | −0.025 (0.031) | 0.033 (0.037) | 0.001 (0.026) |
| Parent age in years | −0.023 (0.639) | −1.284+ (0.643) | 0.529 (0.504) |
| Less than bachelor’s degree | −0.034 (0.032) | 0.012 (0.043) | −0.017 (0.026) |
| Parent rating of letter knowledge (fall) | 0.166* (0.079) | 0.067 (0.113) | 0.017 (0.066) |
| How often child asks to be read to per week (fall) | −0.181+ (0.096) | −0.137 (0.125) | −0.134+ (0.069) |
| How many times per week parent reads for pleasure (fall) | −0.007 (0.081) | 0.129 (0.088) | −0.024 (0.057) |
| How many times per week parent tells a story to child (fall) | −0.043 (0.090) | −0.035 (0.123) | 0.011 (0.068) |
| How many times per week parent sings to child (fall) | −0.007 (0.086) | 0.005 (0.137) | −0.016 (0.062) |
| Child literacy assessment sum score (fall; max. = 126.0) | 3.022 (3.182) | −5.071 (4.322) | −1.502 (2.306) |
| Received texts in English | −0.032 (0.031) | 0.017 (0.040) | 0.025 (0.029) |
| Received texts in Spanish | 0.012 (0.023) | −0.043 (0.026) | −0.015 (0.022) |
| Received texts in Chinese | 0.02 (0.025) | 0.026 (0.032) | −0.01 (0.019) |
| p-value of test of joint significance | 0.410 | 0.016 | 0.320 |
| N | 544 | 251 | 805 |
| Randomization site-by-year fixed effects | ✓ | ✓ | ✓ |
Notes: Fall parent survey responses are standardized. Standard errors are clustered at the randomization site-by-year level. Sample size varies by covariate due to missing data. The sample size indicated in the table is the average number of observations across covariates. The maximum sample size for the parent, teacher, and academic samples are 555, 258, and 821, respectively.
↵+ p < 0.10,
↵* p < 0.05, **p < 0.01.