A test based on panel refreshments for panel conditioning in stated utility measures

Adapting a methodology proposed in Das et al. (2011), this paper uses panel refreshments as a natural experiment to determine whether trends in stated utility measures observed in panel data are genuine or rather caused by measurement issues.