One of the very first investigations of e-dating from an economic perspective is the article “What makes you click: An empirical analysis of online dating (Hitsch, Hortacsu, & Ariely, 2005). The data set for this research recorded the activities of23,000 users in the Boston and San Diego area during a 3 /-month period in 2003.
The data set included self-reported information about users’ age, income, education level, ethnicity, political inclinations, marital status, and so forth. The users also posted pictures ofthemselves on the site, which the researchers rated and ranked (using objective judges) for “attractiveness.” The attractiveness rankings together with self-reported information about users’ height, weight, and other physical characteristics enabled the researchers to create a measure of physical attractiveness that went beyond just the picture.
The most important aspect of this research was that the researchers were provided with information about the activities of the e-daters. Thus, as the authors note (Hitsch, Hortacsu, & Ariely, 2005), “At each moment in time, we know which profile they browse, whether they view a specific photograph, and whether they send or reply to a letter from another user. We also have some limited information on the contents of the e-mails exchanged: in particular, we know whether the users exchanged phone numbers or e-mail addresses” (p. 3).
The researchers based their model on exchange theory, with the obj ective being to assess the impact of some variables (notably, physical attractiveness, but also level of education, income, etc.) on e-daters “success,” as measured by the likelihood oftheir e — mails being responded to and/or by other e-daters agreeing to exchange e-mail address and telephone number with them.
The findings from this research confirmed a number of assumptions that are central to our e-dating model:
1. The percentage of males in the e-dating site (55.5%) was higher than in the general population in both locations (49%) and also higher than the percentage of males among Internet users (also 49%). Other than this demographic difference, all other characteristics of the male and female population of daters were similar to those of the general population and of Internet users in the two cities. Compared to the general population in the two cities, the authors reported that the e-dating users were somewhat younger (25-35), more educated and ofhigher income than the general population.
2. As a rule, the stated preferences ofboth males and females seemed to confirm the findings from the sociological literature reported on in the previous sections that people prefer to date others who are similar to them on all demographic variables. The authors report that this was true for education, income and race, particularly for women.
3. The percentages of people who state that their appearance is “above average” (very high) and those reporting that their appearance is below average (very low) suggests that users “inflate” their reports. This same phenomenon is also evident in regards to weight (much lower than the average for women in the general population) and height (much taller than the average for men in the general population), suggesting that men tend to inflate their height and women tend to deflate their weight in their e-dating profiles.
4. Males and females behaved differently online. Men were more likely to browse women’s profiles than women were (searching behavior). Men were also more likely to send e-mail messages to women than women were, particularly when the men stated that they were interested in a serious relationship. Indeed, this category of men sent more e-mail messages than any other category of e-daters.
5. Males and females were different in the extent to which their actions produced results. Thus, the likelihood of a male to receive a response to an e-mail message was 40%, while the likelihood of a female (irrespective of looks or any other personal attributes) to receive a response to her e-mail message was 70%. The researchers found that attractiveness makes a difference here, with the least attractive women being 2-4 times more likely to send a first contact e-mail message to a man than the most attractive women. The same difference in selectiveness was also evident in the response rate.
6. The final outcomes for daters were that overall, women were browsed more often, received more first contact e-mails and e-mail containing a phone number and/or e-mail address than men. Thus, while men received on average 2.6 first contact e-mails, women received on average 12.6 e-mails; 54% ofall men in the sample NEVER received a first contact e-mail at all, whereas only 19% of all women were never approached by e-mail.
7. As for the impact of self-reported attributes on the likelihood of receiving e-mail messages, it was most pronounced for underweight women (they received 77% more e-mail messages than women who reported that they were overweight) and “blondes.” For men, the “penalty” for overweight was less severe. In contrast, stated income and “shortness” had a strong impact on men’s e-dating “success” while they had a marginal impact on women’s “success.”