Uncontrolled Factors

Description:

The argument draws a conclusion based on a comparison between two (or more) groups, even though some important difference between the groups, other than the difference specified by the experiment, may be responsible for the results obtained.

Comments:

The fallacy applies only to Inductive arguments that draw a conclusion from a special kind of sample in which the sample is divided into distinct groups for purposes of comparison. One such division is into two groups known as an "experimental" group and a "control" group, but other comparisons can also be made, e.g. between two types of medical treatments. Statistics text books often refer to the overlooked factor as a "lurking variable" or a "confounding variable."

Examples:

"We wanted to know whether men or women were more familiar with world affairs. We compared a group of college-educated men to a group of women, many of whom had not gone to college. Our study showed that men were far more likely than women to know the names of significant world leaders and to be familiar with the issues behind foreign conflicts."

We studied two groups of heart patients. Once group received prayers from a local charismatic church; the other group did not. The group that received the prayers, which was the group treated by Dr. Albert, received far fewer pain medications than the other group, which was treated primarily by Dr. Baker. We conclude that prayer helps control pain."

Discussion:

"Controlling" for various factors means making sure that those factors do not affect the results of the experiment. That means that when two groups are being compared, the two groups must be similar to each other in every respect that might be a factor. Of course, it may be impossible to anticipate and control every factor that might be relevant, which is one reason why Inductive arguments can never guarantee a true conclusion, even given reliable observations on representative samples. The fallacy of Uncontrolled Factors can be spotted, however, when a factor that should have been considered was not. In the above example, it should obvious that the attending physician might have an effect on the amont of pain medications that a group of patients receives. Various doctors are known to be more or less conservative in their willingness to prescribe (dangerous and possibly addictive) pain medications.

Source: The term "uncontrolled factors" is widely used by researchers, and any good research methods text book should contain a discussion of the topic. I doubt that an original source for the term could be identified, although the term was probably not used before the 20th Century with the development of research methods in psychology. I believe mine is the first list to include Uncontrolled Factors as a fallacy of Inductive reasoning.

WELCOME                     EXPLANATION OF PRINCIPLES                                     TABLE OF FALLACIES                        EXERCISES                     INDEX