OkCupid Comes Clean About Experimenting On Its Users
John Hopton for redOrbit.com – Your Universe Online
Following recent public outcry regarding experiments that Facebook conducted, in which they manipulated the mood of status updates and other news feed items in order to assess how readers subconsciously reacted in their own subsequent posts, online dating site OkCupid has confessed that it has also been conducting clandestine tests on its subscribers. The admission was not, however, remorseful, and indeed informed the public that they should get used to the idea because “that’s how websites work.”
“Guess what, everybody,” OkCupid’s cofounder Christian Rudder explained in a blog post, “if you use the Internet, you’re the subject of hundreds of experiments at any given time, on every site.”
Debate surrounding this and the Facebook issue has included arguments against the moral outrage which state that manipulating and experimenting on customers is part of what most businesses do whether online or not, and also that such behavior is necessary to improve customer experience and is partly for their own benefit. Despite this fairly sound reasoning, strong reaction to the recent Facebook and OkCupid cases likely comes because the manipulation is of people’s more sensitive emotions.
Facebook, in cooperation with researchers from Cornell University and the University of California, assessed how manipulated feeds affected the activity of unwitting users by removing negative things from the feeds of one test group and positive things from the other. A case for the theory of “emotional contagion” was supported by the outcome, in other words the things that people say on Facebook can be affected, whether negatively or positively, by what they have been viewing from others.
The OkCupid tests were, of course, related to dating. The biggest question was how essential images are in how well somebody is thought of as a potential date. One test in January 2013, called “Love is Blind Day,” involved removing all users’ photos temporarily. Traffic went down, but some people did continue and many of them made faster progress than usual, as chatting got under way more easily and contact info was exchanged sooner. Once photos were restored, though, the chats died off quickly and attractiveness ruled, proving, in the site’s own words, that “people are exactly as shallow as their technology allows them to be.”
Similar tests involved removing or hiding personal information except for photos, and found that personalities of people with attractive photos were still rated highly even with nothing but an image to base the grade on.
“So, your picture is worth that fabled thousand words, but your actual words are worth … almost nothing,” Rudder explained.
Another experiment involved telling daters that they had been matched with “highly compatible” people who were in truth very incompatible by the normal workings of the algorithms, to see if simply believing in compatibility is as important as its genuine existence. It is, according to Rudder, who frankly admits that “the mere myth of compatibility works just as well as the truth.”
As OkCupid makes these unapologetic admissions, the Facebook researchers point out that the investigation was “consistent with Facebook’s Data Use Policy, to which all users agree prior to creating an account on Facebook, constituting informed consent for this research.”
It seems the future may hold more openness about trials such as these, but there will be little or no willingness to reduce their number.