Abstract
We study the effects on results of participants completing a survey more than once, a phenomenon known as farming. Using data from a real social science study as a baseline, three strategies that participants might use to farm are studied by Monte Carlo simulation. Findings show that farming influences survey results and can cause both statistical hypotheses testing Type I (false positive) and Type II (false negative) errors in unpredictable ways.
Original language | English |
---|---|
Journal | Sage Open |
Volume | 3 |
Issue number | 1 |
DOIs | |
Publication status | Published - 16 Jan 2013 |