Lina Abdelfattah

Can Signaling for Potential Intervention Affect Reporting Behavior in Household Surveys?

Cash transfers are more recently common in low- and middle-income countries; one approach to identify who is eligible is to administer household surveys. However, most of these surveys are collected amidst interventions on which respondents may have ex-ante information, so they face the risk of respondents misreporting critical information to impact their eligibility status. Researchers conducting these surveys thus face important trade-off between informing respondents about the purpose of surveys and the associated implications of misreporting. How reliable is self-reported data? Do respondents misreport out of fear of exclusion? Who is more likely to misreport? How does this takeaway assistance from those who ‘truly’ deserve? The effects of awareness about ‘potential’ interventions on reporting behavior of respondents remain understudied, especially in conflict zones where data is scare. To answer these questions, this experiment tests the impact of an explicit signal of a forthcoming intervention in Sudan by randomly varying the placement of such a signal at the beginning or the end of a household survey conducted during war in 2024. No significant differences are found in reported income between treatment groups on average. However, people who don’t perceive themselves as poorer post-conflict, nor poorer compared to their neighborhood, and those impacted by conflict are significantly more likely to misreport. These findings have important policy implications on survey designs and targeting efficiencies, among others, especially in conflict areas where the stakes of even mild exclusion and inclusion errors are typically high.

Agenda > Tous les événements