A number of colleagues and students have asked me to comment on a recent article by Shirley Wang in The Wall Street Journal entitled "The role of deception in scientific research." The type of research we do in PR is based on some of the methods discussed in the article and our studies often fall within the arena of using deception.
Wang began the discussion by offering the example of a professor who studies football hooliganism in Europe. He routinely deceives other football fans by acting as one himself, even engaging in minor criminal acts. This is a form of research called "participant observation." It is routinely used in sociology, anthropology, psychology, and even PR. It is easy to imagine how this research project may enlighten work for a client if that client were a professional soccer team, stadium, league, or city.
But is it deceptive to study football fans by posing as one without telling them they are the subjects of observation? Many researchers say yes. Others say no. So let’s take a deeper dive.
Universities have ethical, thoughtful (and, yes, bureaucratic) boards that review all applications to conduct research with human subjects. Any published research must be reviewed for the protection of those individuals. Public presentation and publication puts research into the public domain rather than the classroom. Therefore, human subjects must be protected – and those protections must include their identities and psyches, as well as their physical well-being.
These review boards protect human subjects, but they do not always rule out the purposeful deception of them. The "Hawthorne effect" exists: if someone knows they are being observed or studied, their behavior changes. Sometimes that behavior change is large, due to a social-desirability bias, "wanting to look good," or artifacts of the study design. Other times behavior changes only slightly. However, it is almost impossible to tell the difference. Add to that mix the fact human subjects need to give permission, called "informed consent," for researchers to include manipulations or specific stimuli and the question becomes very complex.
The intention of a review board is not to stop research, but to ensure it is ethical – that the deception is justifiable or a "necessary condition" of getting data. If the professor mentioned earlier simply did an anonymous survey or conducted interviews about football hooliganism, underreporting of criminal activities or violence would be expected. It is simply human nature. As an ethicist, I teach about the negative aspects of deception on a routine basis. But when is deception justified?
Avoiding Hawthorne effects alone or getting informed consent alone are not reasons enough to warrant deceptive research. There must be no other way to gather data. Additionally, protecting the safety or identity of the researcher is a reason that has been argued as a necessary deception in many studies, especially those in which researchers study potentially dangerous topics or groups. (The football hooligan study is such an example).
This question is an ethical one juxtaposing the deception of participants versus the value of what we can learn as an outcome of that deception. The research should have a redeeming value that could benefit society.
The fracas that ensued when Facebook’s experiment on almost 700,000 people became public is one example of what happens without ethical research standards, not to mention without an oversight board to protect human subjects. Facebook manipulated enormous groups of people, without their consent, to see if certain effects made them happy or sad as a "social contagion."
There was no review board. The subjects were not asked if they wanted to opt into a study. In fact, Facebook stood behind the legalese of its user agreement as a flimsy way of saying that users consented to be studied, even if it is likely the only person who has read the lengthy and banal end-user agreement in its entirety is the attorney who wrote it.
The Proceedings of the National Academy of Sciences published the "massive scale" deception without asking if it was reviewed for human subject protection, or if such a large-scale deception was warranted. Hawthorne effects were not likely to change user behavior and Facebook was not in danger from users, so the deception was not warranted. It was, no doubt, motivated by other concerns, but redeeming value to society was not among them. Neither was human-subject protection. I believe the Facebook study is the Tuskegee experiment of our day. What Facebook did was unethical and it makes conducting research harder for all of us.
Let’s take the opportunity to learn from Facebook’s mistake. Deception can be in ethical research, but knowing when it is absolutely necessary is the crucial consideration. Perhaps our research, and profession, will be better if deception is only an exception to the rule, ethically justified, and warranted.
Shannon Bowen, Ph.D., researches and teaches PR ethics at the University of South Carolina. She is a member of the board of trustees of the Arthur W. Page Society and the board of directors at the International Public Relations Research Conference. Her column will focus on PR education, ethics, and the C-suite. She can be reached at email@example.com.