While meta-analysis is typically used to identify value estimates for benefit transfer, applications also provide insights into the potential influence of methodological design characteristics on results of non-market valuation experiments. In this paper, a meta-analysis of nineteen choice modelling studies in Australia is conducted generating 145 individual value estimates relating to river health. Implicit prices of different measures and scales of river health were transformed into a common standard of willingness to pay (WTP) per kilometer of river in good health. A Tobit model was used to identify the relationships between this dependent variable and a large number of study design characteristics. While there is evidence that the dimensions of choice tasks and description of attributes influence value estimates, there is also evidence that the way tradeoffs and payment mechanisms are framed are equally important. The results of this meta-analysis suggest that more attention should be paid to the way tradeoffs are framed in choice experiments relative to internal choice set structure and data analysis.