ABSTRACT Weighting techniques in web surveys based on no probability schemes are devised to correct biases due to self-selection, undercoverage, and nonresponse. In an interactive panel, 38 survey experts addressed weighting techniques and auxiliary variables in web surveys. Most of them corrected all biases jointly and applied calibration and propensity score adjustments. Although they claimed that sociodemographic and web-related variables are the most useful auxiliary variables to employ in adjustments, they considered only sociodemographic variables to correct biases because of their availability.

中文翻译：

---

使用专家共识（Delphi 方法）评估网络调查中的加权技术，而不是基于概率方案

摘要 基于无概率方案的网络调查中的加权技术旨在纠正由于自我选择、覆盖不足和无响应造成的偏差。在互动小组中，38 位调查专家讨论了网络调查中的加权技术和辅助变量。他们中的大多数联合纠正了所有偏差，并应用了校准和倾向得分调整。尽管他们声称社会人口学和网络相关变量是在调整中最有用的辅助变量，但他们仅考虑社会人口学变量来纠正偏见，因为它们的可用性。