Show Me the Money? Effects of Pre- and Post-Incentive Experiments on Completion Rates in a Statewide ABS Household Enumeration Survey and Subsequent Enrollment in the Full Study
Abstract
For research looking to generalize from a probability-based sample to a larger population, the study’s response rates/completion rates are one of the most important measures of the overall health of the project. This paper focuses on the potential effects of incentive timing (pre vs. promised) and incentive value on completion rates in a statewide ABS household enumeration survey, and then addresses whether the type and value of incentives distributed as a part of the initial household enumeration survey affect subsequent enrollment in the full study.
First: In a statewide survey of New Jersey residents, how much, if at all, does including or not including a visible one-dollar pre-incentive in the invitation letter affect that enumeration survey’s completion rates? Does adjusting the dollar value of the promised post-incentive affect completion rates? Are any incentive-based differences consistent across survey modes (web vs. telephone)? Do incentive-based differences vary by education, household income, race/ethnicity, or immigration status at the census block group level in which a household lies?
Second: For those households who complete the initial enumeration survey, does the incentive timing (pre vs. promised) and incentive value shared during the enumeration stage affect subsequent rates of agreeing to enroll in the full study? Do any effects differ by household composition, such as the presence of immigrants in the household, presence of children in the household, or household size?