Page 266 -
P. 266

254    CHAPTER 9  Ethnography





                           RESEARCHING ONLINE DATING
                           One of the more fascinating topics being addressed by HCI researchers in
                           recent times is the topic of online dating. Individuals go online to various sites
                           (such as http://www.eharmony.com), providing photos and descriptions of their
                           interests in the hopes that they might meet people for dates or relationships.
                           With millions of subscribers of various ages, these sites represent an interesting
                           area for HCI research. Although a number of approaches have been used to
                           study online dating sites, ethnography has not been the primary approach. This
                           raises an interesting question—can ethnography be applied to online dating?
                             At first this might seem like a research focus on individuals but online
                           dating communities are groups with group norms, accepted practices, and
                           shared group communication tools (such as chat rooms). These online groups
                           differ primarily from work groups in terms of the goal of the interaction
                           (dating, not work), the goal of the presentation (to look attractive and
                           interesting, rather than to present information), and the transient population
                           of members in the group (people join and leave the online dating community
                           very rapidly). An example of a group norm and practice is that if you e-mail
                           someone and they do not respond, it is considered totally inappropriate to
                           e-mail them a second time.
                             Hancock et al. (2007) took the approach of recruiting people who
                           were already involved in online dating, to determine the accuracy of their
                           online dating profiles. A self-selected group responded to their recruitment
                           advertisement. The researcher team met with these 80 participants, who
                             presented copies of their online dating profiles (Hancock et al., 2007).
                           Participants were asked to rate the accuracy of their profiles with regard
                           to height, weight, and age. Only 18% of participants had inaccurate age
                           information in their profile but 48% of participants had inaccurate height
                           information and 59% of participants had inaccurate weight information in their
                           online profile. An analysis of the participants' perception of profile accuracy
                           showed that most participants were aware when their profile information was not
                           accurate and were aware that this could be potentially deceptive.
                             Fiore and Donath (2005) examined how people in online dating
                           communities tend to communicate with other people who have similar interests
                           and preferences. The researchers were able to broker an agreement with a
                           dating site to access profiles, statistics, and e-mails (Fiore and Donath, 2005).
                           It is unclear in the paper if users were aware that their profile information
                           was shared with researchers, although it is unlikely (since the researchers did
                           analysis on over 236,000 messages sent from over 29,000 users to over 51,000
                           users). An analysis of 110,000 conversations (messages between a unique
                           pair of users), found that 78% were single messages that were not responded
                           to by the recipient. Users were more likely to contact other users who had
                           similar characteristics (such as “wants children,” smoking, educational level,
   261   262   263   264   265   266   267   268   269   270   271