The Roots of Personality Testing

Source: Kyle Glenn / Unsplash

Personality psychology emerged as an academic discipline in the 1930s. In response to many perceived shortcomings of psychoanalysis and behaviorism, Gordon Allport, one of the founding figures in the field of personality science, described the strengths and weaknesses of the nomothetic and idiographic approaches to personality. These perspectives helped psychology become more scientific by giving rise to empirically testable theories.

But the roots of personality psychology can be traced back to the 4th century B.C.E. at least. Hippocrates, Galen, and other Greek thinkers, were proponents of a bodily-fluid model of personality known as humorism that made a connection between observable behavior and blood, mucus, and bile. We still hang on to the notion of “personality types” despite the original humors—sanguine, melancholic, choleric, and phlegmatic—being completely discredited.

The First Modern Personality Measurement

A century ago, Robert Woodworth developed the Personal Data Sheet (PDS) as a tool to screen United States Army recruits for shell-shock risk. The test consisted of 116 yes-or-no questions that covered somatic, medical, family, and social issues. Published in 1917, the test is widely considered the earliest personality instrument. The administration of the test is straightforward and follows closely the procedures used by many tests today. For example, the questionnaire can be given individually or in groups, and participants are instructed to underline “Yes” or “No” for each question.

The norms for the test were established by giving the questions to two groups of people: normal individuals and a group of known abnormal individuals. Woodworth reported that individuals in the normal groups were likely to answer 10 or fewer questions incorrectly. Answering 30 or more incorrectly was indicative of psychoneurosis.

World War I ended before the test could be implemented for its intended purpose of predicting shell-shock; however, it was used in the early days of personality research as a diagnostic adjunct for assessing “psychoneurotic tendencies” in delinquents. The PDS was later revised in several ways to make it functional with different populations. Some of the original questions rendered the test unsuitable for women and juveniles; the Richmond Modification and the Woodworth-Cady Questionnaire respectively fixed these issues.

Narrow vs. Broad Bandwidth

In contemporary psychological parlance, the PDS exclusively measures emotional stability. As such, it stands out as a prime example of what some researchers would classify as a narrow-bandwidth instrument, measuring just one, two, or at most three traits. Thousands of tests have been designed since the publication of the PDS and most fall into this category. Attributes measured by narrow-bandwidth tests include Optimism, Narcissism, Guilt, and Locus of Control. By contrast, a broad-bandwidth instrument (e.g., MMPI, 16PF, and NEO-PI) captures multiple dimensions of behavior and emotion.

The distinction between the two approaches is critical because while most broad-bandwidth instruments are proprietary, with test items copyrighted by the authors, most narrow bandwidth instruments, including their scoring keys, have been published in scientific journals and books. As part of the public domain, they are freely used by other researchers who can contribute to their further refinement.

There are a number of issues with keeping broad-bandwidth instruments out of the public domain. There is the threat that scientific pursuits may become dominated by commercial interests. Comparative validity studies that would challenge one test against another as predictors of the same crucial benchmarks are not always exhorted. And for many commercially owned inventories, the revisions and changes that would otherwise follow from free and open use by the scientific community are few in number or entirely non-existent.