This thesis contains five studies that make use of non-market valuation techniques and of data collected in Mexico to make methodological and policy contributions to the field. In the order that they are presented in the thesis these are: * a contingent valuation study, based on data collected face-to-face of a representative sample of the population of Mexico City, to calculate a value of statistical life for Mexico and make an assessment of whether the benefit-transfer values that have been and continue to be used in the country are appropriate for policy-making; * a study that uses data collected online on whether the type of organisation sponsoring a contingent valuation survey affects the amount participants say they are willing to pay for the good being valued (in this case mortality risk reductions), all else equal; * a study that uses the same dataset to consider the relationship between trust in institutions and other forms of social capital and contingent valuation results; * an hedonic pricing analysis that makes use of several datasets (including high-resolution property data that is not in the public domain) and seeks to improve on previous attempts at applying this method in a developing country context (jointly using spatial econometrics and an instrumental variables approach); and * a short study on whether there is a relationship between air quality, social capital and subjective wellbeing in Mexico City.