Abstract
This paper uses exogenous variation in the probability of being surveyed at baseline to estimate the impact of being surveyed on subsistence farmers’ take-up of a new agricultural technology that improves food safety. I find large and statistically significant impacts of being surveyed, and also find that an experimental treatment effect disappears for surveyed farmers. My results have strong implications for our understanding of the process of technology adoption, for the external validity of adoption results measured in surveyed populations, and for research ethics.
Original language | English |
---|---|
Pages (from-to) | 941-962 |
Number of pages | 22 |
Journal | Economic Development and Cultural Change |
Volume | 71 |
Issue number | 3 |
DOIs | |
Publication status | Published - Apr-2023 |
Externally published | Yes |