Abstract
This paper reports a novel application of the stated choice method to the valuation of road traffic noise. The innovative context used is that of choice between apartments with different levels of traffic noise, view, sunlight and cost with which respondents would be familiar. Stated choice models were developed on both perceived and objective measures of traffic noise, with the former statistically superior, and an extensive econometric analysis has been conducted to assess the nature and extent of householders' heterogeneity of preferences for noise. This found that random taste variation is appreciable but also identified considerable systematic variation in valuations according to income level, household composition and exposure to noise. Self-selectivity is apparent, whereby those with higher marginal values of noise tend to live in quieter apartments. Sign and reference effects were apparent in the relationship between ratings and objective noise measures, presumably reflecting the non-linear nature of the latter. However, there was no strong support for sign, size or reference effects in the valuations of perceived noise levels. © 2005 Elsevier Ltd. All rights reserved.