Extended Data Table 1 Differential attrition by treatment arms

From: A randomized controlled trial on the effect of administrative burden and information costs on social inequalities in early childcare access in France

  1. Pairwise comparison of answer rates in the endline by assignment group. Column 1 gives the mean answer rate in the relevant comparison group, and Column 2 shows the difference by treatment arm. Models are estimated using OLS with block waves, pairwise fixed effects, and cluster-robust standard errors adjusted at the block wave level. P-values in the table are derived from two-sided t-tests. Padjs (Padj) and confidence intervals account for simultaneous inference using the Holm–Bonferroni correction. Standard errors are cluster-heteroskedasticity robust adjusted at the block level. Standard errors are in parentheses, and 95% confidence intervals are in square brackets. Chi-2 test tests the null hypothesis that all pairwise comparisons are 0. P-values at the bottom of the table indicate the P-value of the Chi-2 test result.
Back to article page