An inventory and analysis has been made of 15 non-market-based valuation studies in Central American and Caribbean countries. The method most frequently used for determining the willingness to pay for drinking water or protected areas was that of contingent valuation. Its main shortcomings were that it was based on open questions, it used contingent scenarios and information frames with a low level of detail, it used small population samples, and it displayed the possible cultural and strategic skews associated with surveys of national residents. The study based on the travel cost method, for its part, was based on low-quality census data rather than on data from surveys of visitors, as well as being based on unrealistic hypotheses regarding the calculation of transport costs, the number of visitors with a single destination, and the levels of surplus consumption of international visitors. The two studies based on the hedonic valuation method were affected by the existence of heterogeneous neighbouring market segments and their inability to obtain reliable ownership values. By way of conclusion, various strategies and a research programme are suggested for improving non-market-based valuation studies in Central American and Caribbean countries.