Conjoint analysis and choice modelling are stated preference techniques for decomposing preferences into the contribution of each of a number of attributes. Both methods can be used for prediction in situations where prices are not observable, and cross-section or time series regression models are not working. Conjoint analysis and choice modelling address these problems by using an experimental design to vary the attribute levels across hypothetical environmental states which can be described to survey respondents. Respondents are being asked a series of questions about their preferences for alternatives environmental states on the basis of selecting the alternative which results in the highest utility. The idea is that an environmental good can be broken down into its attributes. A natural park can be described by the number of trees, the age of the trees, the diversity of its inhabits, diversity of recreational activities. Each of these elements making up a generic natural park is known as an attribute. In a conjoint analysis the respondent ranks a list of combinations of environmental attributes into preference order. Statistical techniques are used to get the relationship between attributes and preferences. If one of the attributes is price, it would be possible to use the resulting preference function to derive the willingness to pay for changes in the levels of the attributes. In a choice modelling study, each respondent evaluates a set of environmental states and indicates which one would be chosen. The alternatives are described in terms of a common set of attributes, called choice sets. The choice sets are differentiated one from the other by the attributes taking on different levels. This is repeated several times for different choice sets. The aggregate 8 choice frequencies can be modelled to infer the relative impact of each attribute level on choice. Respondents’ choices of their preferred alternatives demonstrate their willingness to give up one attribute against another. If one of the attributes used to describe the alternatives has monetary value, it is possible to estimate the respondents’ willingness to pay for additional nonmarket environmental benefits. Tab 4 shows an example of a choice modelling questionnaire.