Item nonresponse is a common problem in contingent valuation research. Discarding incomplete cases will result in biased samples if the discarded cases are nonrandom. Imputing values for missing independent variables and including the incomplete cases may improve overall econometric performance, correct sample bias, and correct bias in aggregate benefit estimates.
No takes yet. Share an insight, caveat, or question.
John C. Whitehead (1994) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context:
More from Journal of Leisure Research