Online Surveys in Nondemocratic Contexts: Evidence from Russia

Abstract How do non-probability online surveys perform in nondemocratic contexts? To address this question, we compared the data from two web surveys (WSs), conducted in Russia in 2018 and 2020, using opt-in panels, to the offline data from wave 7 of the World Values Survey (fieldwork in Russia 2017), round 7 of the Russian Social Survey (fieldwork in Russia 2018–2019), and the 2010 Russian Census data. We found that (a) WSs under-covered older, less educated and rural respondents, compared with both the offline sample and the Census benchmarks; (b) various types of straightlining were somewhat more prevalent in WSs, but the average time spent per item was comparable to that in the offline sample; (c) online respondents expressed considerably more liberal views on such issues as homosexuality, abortion, and divorce, as well as gender equality, compared with offline respondents (by 10% of the total scale range for each domain); and (d) WSs produced slightly different nomological networks of attitudinal orientations.