A test for payment card biases

Abstract

Concerns have been raised that payment cards used in contingent valuation method studies may be subject to range and centering biases. This paper develops a payment card using an exponential response scale. A test for biases is conducted with four versions of the payment card that have greatly different ranges and center values. Range and centering biases are not found except when the payment card truncates, or does not present, the upper end of the value distribution that respondents may desire to select.