Predictions Based on the Actuarial Method

Often predictions based on correlations turn out to be more accurate than judgments rendered by experts. In the influential book Clinical Versus Statistical Prediction (1954) Paul Meehl showed that statistical techniques could do as well as trained experts in predicting a variety of important clinical phenomena, such as the probability that a client would relapse after psychological treatment.

What did Meehl show in his influential 1954 book? What is the actuarial method? What did research show about lie detector experts?

Szucko and Kleinmutz (1981) showed that statistics outperformed expert judges in evaluating polygraph or "lie detector" records. By 1990 Kleinmutz was suggesting that perhaps a combination of using your head and using a formula was best of all. Meehl, still active in the field after 35 years, disagreed. Meehl and co-workers pointed to evidence that actuarial methods-methods using data from the past to predict the future-remained the most accurate. Research showed that when a lie detector expert decided to use common sense to go against predictions based on correlational methods, the expert was usually wrong (Faust, Meehl, & Dawes, 1990).

How can a computer help a psychologist make predictions?

A computer can easily search large data sets for unexpected correlations. The typical result is a statistical formula that out-predicts the experts. That is not necessarily good for the experts. Cohen and Hannigan (1978) told how Cohen was hired to predict who would succeed in a small drug treatment program. Time, money, and energy was wasted on people who dropped out before the program was complete. But nobody could tell which patients would succeed, despite "approximately 60 psychological scores, scales, measures, and quotients" crammed into each patient's folder.

What happened when Cohen got a computer to make incredibly accurate predictions?

After months of struggling with complex clinical data, Cohen got fed up and fed all the information she had into a computer, telling it (in effect) "find anything that correlates with success in the program." To her surprise, the computer quickly located a few variables that predicted success in the program with amazing accuracy. She had accomplished her goal! Now anybody at the clinic could quickly predict which addicts would succeed. So they fired her! Her expertise was no longer needed. Cohen and Hannigan note, "Every silver lining has a cloud."


Write to Dr. Dewey at psywww@gmail.com.

Don't see what you need? Psych Web has over 1,000 pages, so it may be elsewhere on the site. Do a site-specific Google search using the box below.

Custom Search

Copyright © 2007 Russ Dewey