This thesis aims to investigate preferences for Low Emission Buses (LEB). Building on the existing stated preference literature on bus service quality measures and emissions reduction valuation in private vehicles and on the few existing studies looking at LEBs, which have used Contingent Valuation, this study makes use of the multi-attribute nature of Discrete Choice Experiments (DCE) to explore valuations for reductions in two different types of emissions separately, along with other attributes traditionally related to bus travel. This thesis uses a state-of-the-art pivoted experimental design to create reference-dependent choice tasks centred on the respondent's current experience. Choice task realism was additionally sought by using quantitative and qualitative research to inform the experimental and survey design. The survey was conducted at bus stops in the city of Aberdeen during the months of October and November 2016, whilst an existing LEB pilot project was in operation: the Aberdeen Hydrogen Bus Project (AHBP). Choice data is analysed using random parameter logit models to explore preferences for reductions in the two separate emissions and other bus service characteristics. To provide greater insight of the preferences for LEBs, this thesis' additional methodological contributions include using the AHBP as a case study to explore the role of familiarity on preference formation and examining whether choice behaviour is consistent with prospect theory's three main properties: reference dependence, changing marginal sensitivities and preference asymmetry in the form of loss aversion. Policy implications from each of these contribution are discussed.