Table 3.
Changes from Baseline in Student ELA Achievement for Voucher Recipients
| Public to Private with Voucher | Public to Private with Voucher to Public | Private with Voucher to Public | Always Private with Voucher | |
|---|---|---|---|---|
| Pre-baseline achievement level | 0.095*** | 0.009 | — | — |
| (0.016) | (0.034) | |||
| Baseline achievement level | 0.109*** | 0.010 | –0.086 | 0.060 |
| (0.016) | (0.035) | (0.257) | (0.062) | |
| Voucher year 1 | 0.005 | –0.026 | 0.178 | 0.084 |
| (0.011) | (0.024) | (0.135) | (0.047) | |
| Voucher year 2 | –0.011 | –0.130** | –0.105 | 0.115 |
| (0.015) | (0.043) | (0.265) | (0.064) | |
| Voucher year 3 | 0.021 | –0.161* | — | 0.118 |
| (0.020) | (0.081) | (0.102) | ||
| Voucher year 1*structural | 0.032 | –0.018 | — | — |
| (0.020) | (0.037) | |||
| Voucher year 2*structural | 0.006 | –0.038 | — | — |
| (0.023) | (0.064) | |||
| Voucher year 3*structural | –0.015 | –0.200 | — | — |
| (0.032) | (0.137) | |||
| Voucher year 1*nonstructural | –0.008 | –0.030 | — | — |
| (0.014) | (0.031) | |||
| Voucher year 2*nonstructural | –0.023 | –0.170** | — | — |
| (0.020) | (0.054) | |||
| Voucher year 3*nonstructural | 0.046 | –0.143 | — | — |
| (0.026) | (0.100) | |||
| Voucher year 1*grade 4 baseline | –0.057**(s) | –0.036 | 0.245 | –0.075(s) |
| (0.021) | (0.036) | (0.244) | (0.101) | |
| Voucher year 1*grade 5 baseline | –0.030(s) | –0.037 | 0.419 | –0.061(s) |
| (0.022) | (0.038) | (0.253) | (0.105) | |
| Voucher year 1*grade 6 baseline | 0.087***(s) | 0.012 | –0.124 | 0.228**(s) |
| (0.023) | (0.057) | (0.275) | (0.085) | |
| Voucher year 1*grade 7 baseline | 0.057*(s) | — | — | 0.211**(s) |
| (0.024) | (0.079) |
Source: Authors’ compilation based on administrative data provided by the Indiana Department of Education.
Note: Overall results reported from preferred models 1 and 2. To parse out subgroup effects, we include interactions between mobility indicators or grade indicators, and the voucher indicator. We computed separate models by number of years after receiving a voucher and between students starting in public schools versus private schools. ISTEP+ ELA scores measured in standard deviation units, relative to the Indiana statewide mean and standard deviation within each grade and year (across all public and private school test takers). Robust standard errors, adjusted for the clustering of students within schools, are in parentheses. (s)Indicates significant differences between subgroups from F-test of estimates (p ≤ .050). Missing cells indicate l fewer than fifteen students or inability to estimate subgroup differences (for example, no structural/non-structural school transitions for always private voucher students).
*p ≤ .05; **p ≤ .01; ***p ≤ .001