Frequency and Medium Effects on Mean Outcomes
Source: Authors’ analysis based on own data.
Note: Coefficients are from regressions of each outcome, winsorized at the 95th percentile, on a vector of data collection group indicators, randomization stratum fixed effects, and survey week fixed effects (repeated interviews only). Continuous outcomes are standardized to have mean zero and standard deviation one within survey week. Significance tests are based on heteroskedasticity-robust standard errors, clustering by enterprise (repeated interviews only). The lines for each variable show the minimum detectable differences (MDEs) between weekly and monthly in-person interviews in panel (a); the MDEs between weekly in-person and phone interviews are approximately 25 percent smaller. In panel (b) the MDEs are between weekly in-person and phone interviews; the MDEs between weekly and monthly in-person interviews are approximately 5 percent smaller.
