The ethics of deception in social research: A case study

Using data from a study on courtship through personal advertisements, I argue that Kai Erikson's classic case against disguised observation is flawed. Certain kinds of deception are necessary to gather certain data in certain settings. I placed bogus ads in a personal column to obtain and analyze responses. The data would have remained inaccessible—indeed, many of the responses would not have existed in the first place—without some measure of deception. While deception was used, no risk whatsoever was posed to respondents. I further argue that several of Erikson's criteria of risk do not separate ethical from empirical questions; informants use very different criteria in evaluating the risk of harm to them posed by social research that sociologists use. The question of exploitation is more complex, since it has to be weighed against how much of an effort my respondents made and hence, what it is exactly that I took from them. A “panel of judges” decided that most of my male (but not my female) respondents would not have gotten dates with my hypothetical ad placers, and that the research method I used was not especially unethical.