Education and income attraction: an online dating field experiment

ABSTRACT

Prior studies have found a robust correlation in the education of dating and married couples. However, there is little evidence to suggest that such correlations are causal, that is, for the sake of relationship public goods such as the pleasures of the enlightened conversations that only a common high level of education might support. Being empirical studies, they cannot rule out couples matching on other characteristics like income, height or health, which are correlated with education, from driving results. We contribute to this literature by randomly assigning high and low education and income levels to 388 artificial male and female profiles on a large online dating website in China. We then counted thousands of ‘visits’ – clicks on abbreviated profiles, which included education and income information, from search engine results. We found that men’s visits to female profiles were unaffected by the profile’s assigned education level, while women’s visits to male profiles increase with the profile’s education. However, that increase was not increasing on the women’s own level of education, though their visits to the higher income male profiles was increasing on their own education. Our findings suggest that the relationship public goods that stem from a common level of high education are not at the forefront of either men’s or women’s minds before their first dates, when one might expect such goods to play a critical role in the decision to develop the relationship further.

KEYWORDS:

• Education
• online dating
• field experiment
• gender differences
• matching
• marriage

JEL CLASSIFICATION:

• C93
• D10
• J12

Disclosure statement

No potential conflict of interest was reported by the author.

Acknowledgement

The author would like to thank Yan Li and Jue Wang for their excellent RA work, and Jane Zhang and Jue Wang for their helpful comments.

Notes

¹ Such a correlation is not necessarily inconsistent with increased educational homogamy among those who do get married.

² The credibility of the profile is indicated by a positive score, which can be increased by phone verification of the registered phone number, a verifiable ID, uploaded photos, email verification, video verification, a paid membership, etc. Users without a paid membership can browse profiles, while users with a paid membership can, among other things, send first-contact emails to each other. All our profiles have only phone verification and one photo, which gives us the lowest credibility score. However, this score does not appear in search results, and thus, would not affect the observed pattern of visits to our profiles. To affect visits, users would have to search specifically for low credibility profiles.

³ To minimize any possible imposition, we used only profiles which this other website was about to automatically hide because of user inactivity. We are not aware of legal restrictions on the noncommercial use of user created content uploaded to social media websites in China. We assumed that such restrictions, if they exist, are weaker than the United States, where our activities fall under the fair use exemption to the US copyright law. Major US social media websites explicitly announce terms of use that effectively make uploaded user-created content public domain. See, for example, ‘publish content or information using the Public setting’ in https://www.facebook.com/legal/terms. Chinese Universities do not have IRBs to approve the ethics of experiments. However, to the best of our understanding, our design falls under the ‘minimal risk’ exemption from IRB approval. ‘Minimal risk means that the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests’. See here: http://www.virginia.edu/vpr/irb/sbs/resources_regulations_subparta.46.101.html#46.102(i) Those who visit our profiles are not likely to spend more than a few seconds on any of our profiles. Once they visit, they will see no further information. Indeed, we had zero follow-up contact emails from visitors. Visitors are unlikely to encounter more than a few of our profiles since they are spread out among other profiles in any given day and across many days.

⁴ This website does not allow users to report a same sex preference, though users can view anyone else’s profile. Some visitors do not have pictures for their profiles. For these, we can infer the gender of visitors to our male profiles from a feature that was enabled at that time of the male profile treatments. We recorded no same sex visits for them. However, this feature was turned off by the website later, when we did the female profile treatments. We presume but cannot rule out same sex visits from women. Nonetheless, homosexual visitors to our profiles seem unlikely due to the combination of associated stigma, low awareness and the availability of smart phone apps dedicated to homosexual dating.

⁵ These regression results are unchanged if we used years of schooling. We chose the numbers 1–6 because vocational and high school both involve 12 years of schooling, but high school is considered a lower level.

⁶ This model is a simplified version of a standard search model applied to dating and marriage markets. See Hitsch, Hortaçsu, and Ariely (2010b) for a discussion.

⁷ Admittedly, our finding that women’s rates of visits are not increasing on the women’s own education can be explained by more highly educated women believing that it will be harder for them to get a date with more highly educated men than less educated women, because they believe that the men prefer less educated women. However, indifference to women’s education seems the most likely explanation for men’s behavior given that their rates are insignificantly different from constant across all education levels of women for all of their own education levels (‎Observation I). We assume that women’s beliefs about men’s preferences are correct on average.