Air pollution in urban areas is one of the major environmental problems in Chile. In particular, an important number of cities in central and southern Chile exhibit high levels of respirable suspended particulate matter, which are mainly due to emissions from household’s burning of wood for heating and cooking. In fact, the city of Coyhaique in the Chilean Patagonia has been listed as the most polluted city in Latin America (WHO, 2016). Although a number of air pollution control plans have been in place in the last decade, the design of incentive compatible policies towards sustainable use of fuelwood requires the knowledge and understanding of households’ preferences regarding adoption and usage of energy production technologies. This proposal aims at analyzing the determinants of the choice of fuel and the intensity of fuel use for residential heating and cooking in Central and Southern Chile. Because households’ energy production technologies include a variety of fuels, we first investigate households’ choices regarding the use of a particular fuels as their main energy source. We then analyze to which extent the use of fuels relates to fuelwood use, and study intensity of fuel use. By using a unique dataset including a sample of 2,761 households in nine urban areas in central-southern Chile, we propose to estimate multinomial logit models and multinomial logit models with sample selection to take account of the choices of fuelwood users. Moreover, we estimate both ordered and count models in order to analyze intensity of use and households’ energy profiles. The knowledge of these factors brings an opportunity for the design of future policy interventions aimed at incentivizing the adoption of cleaner devices.