US Environmental Protection Agency, Office of Research and Development, 26 West Martin Luther King Dr. (MS 587), Cincinnati, OH, 45268, USA.
Appalachian State University, Department of Economics, 416 Howard Street, ASU Box 32051, Boone, NC, 28608, USA.
J Environ Manage. 2024 Feb 14;352:119829. doi: 10.1016/j.jenvman.2023.119829. Epub 2024 Jan 6.
This study quantitatively reviews the hedonic literature examining surface water quality to assess how attributes of the commodity, housing market, and methodological choices lead to variation in the significance and expected sign of the estimated property value effects (i.e., elasticities). We conduct a meta-analysis of 29 studies with 290 unique estimates, published or released between 1985 and 2017, and find evidence based on probit meta-regression models that some of the definitions and decisions made in primary studies do influence the estimated relationship between water quality and home prices. Our most robust evidence suggests that methodological choices (e.g., accounting for spatial dependence, or if the water quality measure was based on something other than in situ measurement) have a critical role in determining the likelihood of finding a significant and theoretically expected result; and perhaps most importantly, it is not always selections that reflect best practices that lead to this finding. This study can help identify potential concerns with data and modeling choices in the collective hedonic literature focused on water quality.
本研究对考察地表水水质的享乐主义文献进行了定量回顾，以评估商品、住房市场的属性以及方法选择如何导致估计的财产价值效应（即弹性）的显著程度和预期符号的变化。我们对 1985 年至 2017 年间发表或公布的 29 项研究中的 290 个独特估计值进行了荟萃分析，并基于概率单位元回归模型发现，一些主要研究中做出的定义和决策确实会影响水质与房价之间的估计关系。我们最有力的证据表明，方法选择（例如，是否考虑空间依赖性，或者水质衡量标准是否基于原位测量以外的其他标准）在确定是否有可能找到具有显著意义和理论预期结果的关系方面起着关键作用；也许最重要的是，导致这一发现的不一定是反映最佳实践的数据和模型选择。这项研究有助于识别集中于水质的整体享乐主义文献中数据和建模选择方面的潜在问题。
