Do OkCupid and Facebook experiment on vulnerable populations?


Keisha Ray

Publish date

Tag(s): Legacy post
Topic(s): Informed Consent Media Research Ethics

by Keisha Ray, Ph.D.

A few months ago Facebook announced that some Facebook users were a part of a 2012 experiment. In the experiment Facebook altered the number of negative and positive posts and photos that appeared in users’ newsfeed. In a paper documenting the results of the study, authors noted that by changing what users saw in their feed, Facebook was able to alter moods, emotions, and the kind of posts that people posted. The study was meant to be an experiment in online social interactions and emotional connections.

OkCupid, an online dating site has also recently announced that it has conducted experiments on its date seeking users. The experiments ranged from altering information to determine its affect on personality ratings to being less than truthful about the strength of users matches with other users. Making these changes to profiles and users’ information, researchers were able to determine the likelihood of interactions between users and the mood of those interactions.

There are many reasons to object to the kinds of experiments conducted by Facebook and OkCupid. We can object on the grounds of being misled. Facebook and OkCupid led users to believe that the information given to them has not been manipulated. They were led to believe that the posts, photos, and profiles that appear on their home pages have not been modified, but they have been. Being misled is enough to make users angry. But that still leaves questions about why being misled ought to make users angry. A source for outrage over Facebook and OkCupid experiments is that the information on our screens is tied to our emotions. Posts, photos, and profiles on sites like Facebook and OkCupid are tied to the way that we see ourselves, the selves that we want other people to see, our happiness, and our sadness. If companies manipulate our emotions to gather data on social interactions then in essence, they are in control of our emotions. We like to think that our emotions are our own, that they are personal and almost a sacred extension of our brain. To control our emotions is to control our minds. Controlling our minds likens us to lab rats who serve the instrumental purpose of data collection. But people are supposed to be treated with respect. Our ability to reason and feel pain and pleasure is supposed to be the basis for that respect, but experiments that test users’ emotions treats people as less than such.


Vulnerable populations
Considering that our emotions are tied to our online social interactions and our emotions were the subject of experimentation, we have to consider the ethical nature of this kind of research. Research of this kind may be legal if the possibility of using users’ data in this manner was disclosed in a “terms of use” agreement that all users must agree to for access to the sites’ services. However, it can be unethical on the grounds that the research was misleading and manipulative in the ways that experiments on vulnerable populations are manipulative, and therefore, unethical.


Vulnerable populations are typically described as those that cannot decline participation in experiments either because they have not been informed of the experiment or they are in a subordinate position as compared to those conducting the experiment. In both experiments conducted by Facebook and OkCupid, users were unable to opt out of the research because they were unaware that it was taking place. Being informed of the experiment and given the opportunity to opt out would have likely jeopardized the integrity and usefulness of the acquired data.

If you are not convinced that Facebook users and OkCupid users can be considered vulnerable populations, we can consider the reasons why people use sites such as OkCupid. People who use dating sites are looking for a service and sometimes pay for that service. Assuming that at least some OkCupid users are searching for meaningful romantic relationships and this search requires emotions, OkCupid is in the business of emotion. OkCupid assesses users personalities to match them with other users whose personalities’ questionnaires reveal a good match, while also revealing those users who are not a good match. These assessments are intended to spur conversations and these conversations are intended to create meet ups and dates. However, if conversations, meet ups, and dates are arranged based on manipulated assessments and manipulated “matches” then OkCupid is creating connections based on manipulated emotions. If this is the case, then OkCupid is showing a great amount of disregard for the emotions of those searching for meaningful relationships. OkCupid users are looking for basic elements to a good life—relationships (of all kinds)—and despite knowing this and knowing that relationships are tied to emotions, OkCupid used those emotions against its users to acquire data on social interactions, making data more important than its users.

It is true that social networking users are not a vulnerable population in the same ways that certain cultures, races, and socioeconomic and geographic groups are vulnerable populations. However, social networking users, in particular users of online dating sites like OkCupid, are similar to vulnerable populations in that they are a group that can easily be manipulated and whose health can be affected by such manipulation. The ease of this manipulation is demonstrated by the lack of knowledge of these experiments once they were revealed to the public and the public’s shock and discomfort with the experiments.

We use cookies to improve your website experience. To learn about our use of cookies and how you can manage your cookie settings, please see our Privacy Policy. By closing this message, you are consenting to our use of cookies.