The Effect of Attribute Representation Methods on Noise Valuation: A Choice Experiment Study

Kaushali Dave,Jeremy Toner,Haibo Chen
DOI: https://doi.org/10.1016/j.trd.2018.01.011
2018-01-01
Abstract:Traffic noise has been known to severely affect human population. The valuation of traffic noise pose a significant challenge in choice experiments as respondents have little understanding of the physical measure of noise and its associated perception. As a result, several techniques have been developed that represent noise using different methods, either based on the level of noise exposure or the respondent's level of noise annoyance. This study examines the effect of different methods of attribute representation on respondents' attribute understanding and valuation. The study is focussed on residential choice and residential view and sunlight are important attributes that are examined along with traffic noise. The study demonstrates that the methods of attribute representation have an important effect on respondents' understanding of the attributes as well as in the subsequent valuation. It was found that attribute such as view is better represented using the location representation while noise is better represented using the linguistic method. Moreover, the method of attribute modelling also plays a significant role in the analysis as certain data input techniques are more suitable for some representation methods.
What problem does this paper attempt to address?