Data from the experimental project 'Love or politics? Political views regarding the war in Ukraine in an online dating experiment' by Anna Beloborodova
Description
This dataset contains data from the experiment and python code for the project titled “Love or politics? Political views regarding the war in Ukraine in an online dating experiment”. Paper abstract: How polarized is Russian society regarding the war in Ukraine? Political views affect various behaviors, including relationship formation. This study conducts a field experiment on a large Russian dating site and gathers data from over 3,000 profile evaluations. The findings reveal significant penalties for those who express pro-war or anti-war positions on their dating profiles. Age emerges as the most polarizing factor: younger individuals are less likely to approach pro-war profiles but not anti-war ones, whereas older individuals are less likely to respond positively to profiles indicating anti-war views but not pro-war ones. The results align with survey evidence of a positive relationship between respondents' age and expressed support for the war in Russia, although the experiment indicates a higher degree of polarization. Overall, the experimental findings demonstrate that survey data can reveal trends and relationships between individuals' characteristics and their opinions, but may overstate the levels of support for government agendas in non-democratic states. The experiment was conducted in October - November, 2022, on a large online dating site in Russia in three Russian regions: Moscow, Saint Petersburg, and Sverdlovskaya oblast. There are three separate data files, one for each region. Each file contains information on dating site users that have been liked by and/or have viewed the experimental profiles. The data have been updated and can be found in the DataAndPython-FSravl.zip file. Here are the changes that have been made: 1) One user's ID typo has been corrected. 2) File ExperimentDataMainLikedUsers.csv contains data on the main sample of liked users. The hair color of these users was recorded from profile photos whenever possible. Weights have also been added to enable analysis with adjustment for differences in age distribution between dating site users and a subset of the Russian population that shares similar observable characteristics. The folder also contains python code for data analysis. The description of the study is available at https://mpra.ub.uni-muenchen.de/118862/ .