Chapters 9 through 11 described methods for inferring individuals’ values for environmental amenities from their observed choices of related market goods. These approaches are broadly categorized as “revealed preference” (RP) techniques because, it is argued, a person’s actions in the marketplace reveal information about his core preferences, including his preferences for public goods. 1 However, discussions in Chapter 4 also showed that there are many circumstances under which value measures cannot be derived from market transactions. This is the case, for example, when individuals are thought to place value in an environmental amenity that they do not directly use (that is, so-called “passive-use” values). Because such values are not tied to the use of a related market commodity, they leave no footprint in the marketplace from which their magnitude can be inferred. Even in the case of use values, revealed preference techniques may be hampered by a lack of sufficient independent variation in the amenity of interest from which to infer its impact on behavior and preferences. In the context of recreation demand models, for example, it may be difficult to discern the marginal value associated with a specific site amenity (e.g., a reduction in lake algae) when there is limited variation in the amenity across existing sites, or a high degree of correlation with other site attributes (e.g., fish catch rates).