Wednesday, March 5, 2008

New MMPI scale invalid as forensic lie detector, courts rule

Injured plaintiffs falsely branded malingerers?

Psychology's most widely used personality test, the MMPI, jumped into the national spotlight today in a fascinating David-and-Goliath controversy pitting corporate interests such as Halliburton against the proverbial little guy.

At issue is the "Fake Bad Scale" that was incorporated into the Minnesota Multiphasic Personality Inventory last year for use in personal injury litigation. A front-page critique in today's Wall Street Journal includes publication of the items on the contested scale, a test security breach that will no doubt have the publisher seeing red.

Although a majority of forensic neuropsychologists said in a recent survey that they use the scale, critics say it brands too many people - especially women - as liars. Research finding an unacceptably large false-positive rate includes a large-scale study by MMPI expert James Butcher, who found that the scale classified high percentages of bonafide psychiatric inpatients as fakers.

One possible reason for this is that the scale includes many items that people with true pain or trauma-induced disorders might endorse, such as "My sleep is fitful and disturbed" and "I have nightmares every few nights." Yet hearing the term "Fake Bad" will likely make a prejudicial impact on jurors even if they hear from opposing experts who say a plaintiff is not faking.

The controversy came to a head last year in two Florida courtrooms, where judges barred use of the scale after special hearings on its scientific validity. In a case being brought against a petroleum company, a judge ruled that there was "no hard medical science to support the use of this scale to predict truthfulness.” Other recent cases in which the scale has been contested include one against Halliburton brought by a former truck driver in Iraq.

The 43-item scale was developed by psychologist Paul Lees-Haley, who works mainly for defendants in personal injury cases and charges $600 an hour for his depositions and court appearances, according to the Journal article. In 1991, he paid to have an article supportive of the scale published in Psychological Reports, which the WSJ describes as "a small Montana-based medical journal."

The scale was not officially incorporated into the MMPI until last year, after a panel of experts convened by the University of Minnesota Press reported that it was supported by a "preponderance of the current literature." Critics maintain that the review process was biased: At least 10 of the 19 studies considered were done by Lees-Haley or other insurance defense psychologists, while 21 other studies – including Butcher's – were allegedly excluded from consideration.

Later last year, the American Psychological Association's committee on disabilities protested to the publisher that the scale had been added to the MMPI prematurely.

Lees-Haley, meanwhile, defends the scale as empirically validated and says criticism is being orchestrated by plaintiff's attorneys such as Dorothy Clay Sims, who has written guides on how to challenge the Fake Bad scale in court.

Even if the scale was valid before today, questions are certain to arise about the extent to which it will remain valid once litigants start studying for it by using today's publication of all 43 items along with the scoring key.

The lesson for forensic practitioners: Be aware of critical literature and controversy surrounding any test that you use in a forensic context, and be prepared to defend your use of the test in court.

The article, "Malingerer Test Roils Personal-Injury Law; 'Fake Bad Scale' Bars Real Victims, Its Critics Contend," which includes ample details on the controversy, is only available to Wall Street Journal subscribers, but you can try retrieving it with a Google news search using the term "MMPI Fake Bad." The University of Minnesota Press webpage on the contested scale is here, along with a list of research citations.

Here are citations to the major pro and con research articles:

"Meta-analysis of the MMPI-2 Fake Bad Scale: Utility in forensic practice," Nelson, Nathaniel W., Sweet, Jerry J., & Demakis, George J., Clinical Neuropsychologist, Vol 20(1), Feb 2006, pp. 39-58

"The construct validity of the Lees-Haley Fake Bad Scale: Does this measure somatic malingering and feigned emotional distress?: Butcher, James N., Arbisi, Paul A., & Atlis, Mera M., Archives of Clinical Neuropsychology, Vol 18(5), Jul 2003, pp. 473-485.

Postscript: Test distributor Pearson Assessments responded with alacrity - not to the heart of the controversy but to the Journal's reprinting of test items. The company, which
makes a mint from selling and scoring the MMPI and other psychological tests,got the WSJ to remove the online link to the test items. In a "news flash," Pearson says it is "evaluating the impact of the article" and asks psychologists to report any other instances of "illegal" reproduction of the scale in publications, websites, chat rooms, or blogs.

NOTE: For more of my posts about the MMPI-2's Fake Bad Scale, search the blog using the term "MMPI" (the search box is in the upper left corner of the page).

3 comments:

  1. This is a very worrying development.

    Without seeing the items used it is very hard to evaluate the validity of the comments that have been made.

    On the face of it it seems very odd that responsible and experienced researchers, practitioners and publisher would give creedance to so manifestly biased and confused an instrument.

    I would be very interested to have links to further research evidence and arguments.

    jeff hughes, psychologist.co.uk

    ReplyDelete
  2. yes, it is odd that researchers would give creedance to so 'confused an instrument;'. but the words experienced and responsible are BIG assumptions. remove those and the pool clears. This is an industry of recent unethical nazi affiliated experiments as recent as the seventies. see project paper clip. very nasty. nazis under the bed

    ReplyDelete
  3. Fortunately, I seem to be able to score however I want on the MMPI2, but had trouble with the "fake bad" when I was trying to get my S score down.
    What a silly business.

    ReplyDelete

 
Real Time Web Analytics