Results of choice experiment studies are widely claimed to provide valuable inputs into more efficient environmental policy development. The implicit price estimates for the attributes included in the choice experiment give policy makers an indication of the non-market values of environmental goods and services. There are, however, few standardised approaches to guide the choice of the environmental attributes. Although recent publications (Boyd and Krupnick, 2009; Johnston and Russell, 2011) stress the need to base the definition of non-market environmental attributes in ecological theory, choice experiment studies often give minimal evidence to support the choice of attributes. This paper reviews ten years’ worth of choice experiment studies in leading environmental economics journals. The aim of this study is to investigate on what basis the attributes and units used in the valuation studies were chosen, and how the survey development process is reported. The review shows that only very few published papers report the evidence sources on which the choice of attributes and their levels was based. The disjoint between evidence-based research method and the reporting of protocols in choice experiment valuation studies undermines the credibility of nonmarket value estimates to people outside the profession. There is a need for greater attention to transparent, evidence-based survey development to support more robust welfare estimates and withstand criticism.