Illusory association happens when an association between two attributes is not direct; The association is dependent on a third, hidden or unrecorded attribute.
“What one Christian does is his own responsibility, what one Jew does is thrown back at all Jews.” Anne Frank, May 22, 1944 (Frank, 1993, p. 239)
Major crimes (like murder or robbery) are relatively rare. When a minority commits one of these crimes (i.e. a member of a relatively rare group), it’s likely to be more memorable than if the crime was committed by a member of the majority. This creates an illusory association between severe crime and minority groups (Hamilton and Gifford, 1976).
Causes for Illusory Associations
Illusory associations can happen for many reasons, including:
- Sloppy research: Let’s say a researcher’s concentration wanes while studying characteristics X and Y. He only notices X when Y happens (Y could be a loud noise or a similar attention-grabbing feature). By only noticing X when Y happens, the researcher has created an illusory association between X and Y, without considering Z: the researcher’s attention.
- Biased researchers: sometimes a researcher might be more inclined to record presence of a characteristic, rather than absence of one, perhaps because of personal bias.
- Improper sampling: if you don’t sample correctly, you can’t rely on any results in your study—including any apparent associations.
Sometimes illusory associations happen even in carefully designed research projects, if attributes are not distinct and data is misclassified.
Avoiding Illusory Associations in Research
We defined illusory association as the association between two attributes which have no real connection other than their mutual connection with a third factor, which was ignored. Avoiding illusory associations is as simple as giving proper acknowledgement to the connecting factor.
In practice, this involves a careful survey of the attributes and other related attributes which might play the role of ‘hidden middleman’.
Agarwal, B. (2006). Basic Statistics. New Age International.
Frank, A. (1993). The diary of a young girl (B.M. Mooyaart-Doubleday, Trans.). New York, Bantam.
Hamilton, D. & Gifford, R. (1976). Illusory correlation in interpersonal perception: A conitive bias of stereotypic judgments. In Journal of Experimental Psychology. 12, 392-407.
Yule, G. (1911). An Introduction to the Theory of Statistics. C. Griffin.
If you prefer an online interactive environment to learn R and statistics, this free R Tutorial by Datacamp is a great way to get started. If you're are somewhat comfortable with R and are interested in going deeper into Statistics, try this Statistics with R track.Comments? Need to post a correction? Please post on our Facebook page.