Partial attribute attendance in environmental choice experiments: A comparative case study between Guangzhou (China) and Brussels (Belgium).

摘要: Attribute non-attendance (ANA) in discrete choice experiment (DCE) exercises has attracted increasing, yet limited, scholarly attention. This paper attempts to investigate ANA a comparative case study, with focus on its patterns and their association socioeconomic, behavioral perceptual factors, as well impacts willingness-to-pay (WTP) estimates. We deploy four-level polytomous scale (always, often, seldom, never considered) for respondents state various degrees of attribute attendance (SANA) an identical DCE questionnaire about urban river restoration initiatives two global cities contrast socioeconomic contexts, similar request restoring polluted modified rivers, Guangzhou (south China) Brussels (Belgium). The survey results reveal the existence large proportions partial sampled cities. use extended mixed logit model, which incorporates separate parameters delineating each attribute's different groups, estimate respondents' average WTP values. find that accounting SANA could improve goodness-of-fit model affect magnitude mean Respondents' level pertaining attributes is mainly associated perceived importance rivers' ecosystem services, but may not be necessarily correlated strength preference corresponding indicated by Whether this discontinuity between stated levels estimates within sample questions ability DCEs generate unbiased welfare estimation policy guidance developing countries calls further studies.