The Advantages & Disadvantages of Field Experiments in Sociology

In his book, “After Virtue,” philosopher Alasdair MacIntyre said that social sciences have consistently lacked predictive power because they are incapable of articulating law-like generalizations in the same way that physical sciences do. The ability of humans to invent, decide and reflexively react, as well as plain chance, makes any field experiment in sociology a risky proposition.

A sociological field experiment is any experiment that is carried out beyond the laboratory; that is, the experiment is performed in a “natural setting,” by observation of an unprepared environment, or the experiment is performed in an artificial setting where the variables cannot be controlled. The fact that variables cannot be controlled is the major disadvantage of field experiments from the scientific standpoint. In laboratory experiments, the control of variables makes it possible to establish relations of cause and effect.

An advantage of field experiments is that the subjects are presumably not as influenced by the observations of the experimenters -- especially if they do not know an experiment is taking place. An experiment sending white and black actors into a similar social situation is one example of subjects who are unaware that an experiment is being conducted. Had they been aware of being subjects, it is unlikely they would have acted naturally -- in this case behaving differently toward black and white actors who otherwise had the same attributes.

Sociological field experiments present researchers with significant ethical problems. Because field experiments may lack a strong element of control, there is a higher risk of unanticipated actions that can adversely affect subjects and participants. In the case of the Stanford prison experiment, a simulated prison was established to study interactions between students who had volunteered to play the roles of prisoners and guards. Within days, the “guards” had unexpectedly become so abusive that there was serious risk of injury or death to prisoners, and the experiment was shut down. Much of value has been inferred about the behavioral power of roles from this experiment, but there are serious ethical questions about whether researchers have a right to create this kind of risk for subjects.

Some people might agree or disagree with the statement, "Most people are honest." Some field experiments that can be done simply and at low risk can be used to prove or disprove popular assumptions about social behaviors. A good example is the “honest tea” experiment. A tea company set up unmanned stands in seven cities with bottles of tea for sale at a dollar apiece. Customers were instructed to put a dollar in the money box for a bottle of tea. Hidden cameras recorded that between 75 percent (in Los Angeles) and 93 percent (in Boston) were honest and paid. Obviously, there may have been variables in each city that affected the differences, so no conclusion about the cities are valid; but a general conclusion might be that most people in this situation are honest.