Gordon, R.M.  (2006d) An Expert Look at Love, Intimacy and Personal Growth. IAPT Press, Allentown, Pa.  (chapter 3 Measuring Individual Traits)

Chapter 3  Measuring Individual Traits

One of the first things that a scientist does is to classify and measure. A good way to measure personality is with objective psychological tests. "How to Pick a Good Apple" was written by Debra Kay Woolever Bennett, and me for Security Management (1986) to explain in lay terms the MMPI and its value for security screening.  

A recent report, “The Insider Study,” issued by the US Nuclear Regulatory Commission (NRC) reported that of eleven incidents involving a potential threat to public safety during a three-year period, all were due to inside employees.

In its continued interest to learn about the best way to hire the right employees, the NRC commissioned a study. Psychologists F.D. Frank, B.S. Lindley, and R A. Cohen found the Minnesota Multiphasic Personality Inventory (MMPI) superior to the polygraph, personal interviews, other psychological tests, and reference checks or recommendations when the criteria for tests included validity, reliability, compliance with legal issues, employee selection procedures, personal effects on the applicant, and susceptibility to faking.

The MMPI is a true/false questionnaire that measures many components of personality and is considered the best objective tool for assessing psychopathology. The MMPI has over 10,000 research studies spanning over forty years, including norms based on 50,000 medical patients from the Mayo Clinic.

MMPI scores are based on how individuals choose to respond to a pattern of items in comparison to how a known diagnostic group responds. The test was not constructed based on how psychologists thought people would react to different items. For example, psychopaths respond to certain questions on the Psychopathic Deviate Scale of the MMPI in a characteristic way -- they tend to deny things about themselves in a manner typical of psychopaths. The MMPI established norms for psychopaths and other groups.

Dahlstrom, Welsh, and Dahlstrom (1972) concluded that the MMPI has “proven to be more dependable across situations and across patient populations than human judges with different levels of training, different diagnostic philosophies, and different kinds of clinical experiences.”

I (RMG) saw the potential for using computerized MMPI reports in 1980 when the microcomputer became available. I developed reports that would generate objective decision rules for classifying test profiles as "pass", "fail", or "indeterminate" security risk.

After investigating the best-decision rules to utilize in the report, I tested the predictive validity of these rules by comparing them to actual case studies. I administered my MMPI report to fifty-two outpatients.

The MMPI authors, Hathaway and McKinley developed the MMPI for psychiatric inpatients. The original clinical scales are, Hypochondriasis, Depression, Hysteria, Psychopathic-Deviate, Masculinity-Femininity, Paranoia, Psychasthenia, Schizophrenia, Hypomania and Social Introversion.

In my computerized report I, added additional scales related to security risk. These scales measure: threat of suicide, violence, hostility, addiction proneness, blocking evidence of personal error, authority conflicts, resisting being told what to do, and poor work attitudes.

I also included positive personality factors such as intellectual efficiency, ability to tolerate confrontations, and self-sufficiency.

All testees were interviewed and rated on a scale of 0 to 3 where 0 indicated no risk and 3 indicated a definitive security risk.

The results indicated that all people considered a security risk were detected -- eight out of fifty-two (100 percent hit rate). The decision rules also passed everyone who should have been passed. The decision rules produced an overall hit rate of 84 percent.

It is important not to fail someone just because of any psychopathology. However, indications of psychotic traits (poor perception of reality, confused thinking, delusions, poor emotional controls) and psychopathic traits (poor impulse control, hostility, addiction proneness, poor judgment, irresponsibility) are clear bases for a failure.

Assuming that a failure is any sign of psychopathology (any elevated MMPI clinical scale or indication of faking to look good or bad), approximately 50 percent of laborers and 30 percent of office workers would fail according to a sample of 3300 nuclear plant workers. This would give too high a fail rate (too many false positives). After a clinical interview, a final fail rate should range between 1 and 5 percent.

I and Drs. Baird, Gerdes, and Martenis found that 132 out of 3300 (4 percent) nuclear plant workers failed both the MMPI screening and subsequent interviews.

This showed that although many people experience emotional problems, few people could be considered potentially dangerous.

Personality states are brief. In contrast, personality traits result from a combination of inherited temperament and childhood relationships with parents. These personality traits usually endure for a lifetime. The MMPI measures personality traits that cause personal and interpersonal distress and impairment. The MMPI in combination with an interview is valuable in detecting possible personality traits that are associated with poor judgment, mood disorders, impaired reality testing, authority conflicts, hostility, paranoia, impulsiveness and addiction proneness. Although it is hard to predict who will be dangerous, the combination of an MMPI and a clinical interview can scientifically narrow the field and help to select reliable staff.

However, assessment does not stop at a single testing. The management needs to have an ongoing recognition of emotional and stress problems in others. Supervisors will need to refer workers for assessment when they suspect an escalation of symptoms over time.