ABSTRACT
In considering the assessment of validity, we are guided by two assumptions. First, not all tests of validity are of equal importance. We will devote special attention to experiments which have compared hypothetical and real-payment markets for the same good, on the premise that the realpayment or "simulated" markets serve as a particularly useful criterion against which to judge the accuracy of hypothetical scenarios. Second, no single test is definitive. Both the nature of public goods and the multidimensional character of validity conspire against the results of any test "proving" that the method is a valid way to measure benefits. Moreover, even if a particular CV study is found to be invalid, the reason may lie in its poor design rather than in the method itself. Our objective in this chapter is to examine the overall evidence of validity.
