Face-Lift for a Famous Test

By John Leo

Chicago Psychiatrist Sidney Weissman derisively calls it "the old Sears catalog" of psychological tests. The Minnesota Multiphasic Personality Inventory is indeed one of the oldest, longest and most cumbersome tests in use today. Millions of people in at least 46 countries, from psychotics to normal job seekers to Soviet cosmonauts, have puzzled their way through its seemingly endless array of odd and eerie statements (samples: "Much of the time my head seems to hurt all over"; "My soul sometimes leaves my body"; "In walking, I am very careful to step over sidewalk cracks"). Now, at age 44, the archetypal test is getting a face-lift. "This revision is long overdue," says Kent State Psychologist John Graham, one of four professors who have been working on the test for four years. "We psychologists have been less than responsible in letting it go so long."

The MMPI, which grew out of work in the late 1930s at the University of Minnesota by Psychologist Starke Hathaway and Psychiatrist J.C. McKinley, assesses character, attitudes and behavior by the patterns of responses (true, false, cannot say) to 566 statements. Originally designed to help identify those with psychological problems, the test was widely used in clinical settings during World War II by military officials who wanted quick findings about attitudes toward authority, impulse control, drive to dominate and other potentially troubling aspects of the normal personality.

Some revisions are minor or obvious. Dated references to streetcars, sleeping powders and the children's game drop the handkerchief will come out. The statement "I like to take a bath," which had been attracting "false" answers from shower lovers, will become "I like to take a bath or a shower."

Other changes will be more complex. As the MMPI has come to be seen as a beloved landmark of American psychology, it has also come under frequent attack as dated and culture bound. Since empirical work on the test was done among pre-war, white, rural Minnesotans in their mid-30s, it does not account for newer values and is often a particularly unreliable test for blacks, women and adolescents. On the masculinity-femininity scale, a woman who says "true" to "I would like to be a soldier" or "I like mechanics magazines" risks being pigeonholed as abnormally masculine. The test uses "he" instead of "he or she," and over the years, some researchers have noted that women have complained about the test far more frequently than men.

Many psychologists contend that the test does not work well with blacks. One study found that blacks and whites give significantly different responses on 213 of the 566 items. By 1960, with the test nearly two decades old, only four of the approximately 1,000 articles on the MMPI discussed its impact on blacks, and those four were confined to criminal and psychiatric populations. The test also needs better adaptation to the psychology of adolescence. For almost 40 years, some psychologists have noted that the MMPI profile of the normal youngster temporarily caught in adolescent turmoil is similar to that of the adult psychopath.

The revision committee is now testing two experimental booklets, one for adults, one for adolescents. Some 15,000 Americans and Canadians, randomly selected from phone books and replies to magazine ads in eight states and the city of Toronto, have taken the new forms of the test. Unlike the original sample -- now regarded as "both small and parochial," according to Committee Member James Butcher of the University of Minnesota -- the new group is carefully balanced by region, ethnic group, age, education and gender. Though the committee members decline to talk in detail about the revisions or to release copies of the test booklets, they say they have kept the old test virtually intact and added some 150 items.

Psychologist Graham says the new items are designed "to measure things we didn't think the MMPI measured very well," among them the possibility that therapy will help the person tested. Such an option was not thought of in the '30s, he said, because psychologists of the day merely diagnosed and rarely treated the disturbed. Many additions deal with fresh concerns, such as eating disorders, drug abuse and the Type A personality. One thing that will not change: the simple, almost simpleminded prose. Like the original, the language of the revised version is aimed at a sixth-grade level, though Americans are vastly better educated now. According to Committee Member Grant Dahlstrom of the University of North Carolina, "People don't read as much as they used to. We are concerned that we don't raise the reading level too high." Nonetheless, he says, the MMPI will remain an excellent diagnostic tool: "It's still pretty hard to fake it."

Psychologists seem to regard the revision as necessary, but some are wary of abandoning old standards. George Taylor of Atlanta, president of the Georgia Psychological Association, says he has serious reservations about establishing new definitions of acceptable behavior: "It's not clear how the new norms will relate to the old norms, which generated a huge body of literature. How useful would that material be? It's like starting all over again." Guy Seymour, a psychologist for the Atlanta police bureau, calls the updating "wise and timely" but predicts that it will have little effect on the comments of police applicants who take the test. The average reaction, he says, is likely to remain complete bafflement over items like "I like tall women."

With reporting by Michele Donley/Chicago and Don Winbush/Atlanta