What are personality rating scales?

Introduction

Personality rating scales represent one approach that is used by psychologists and others to measure scientifically dimensions of personality for purposes of summarizing, predicting, and explaining human behavior. In recent history, there has been considerable research on and application of the use of personality rating scales, typically to measure psychological individual difference dimensions or traits along which people can be ordered, such as extroversion or neuroticism. Traits are consistent patterns in the way individuals behave, feel, and think.

Trait psychology can be considered the theoretical underpinnings for measurements of personality, including personality rating scales. However, the use of rating scales does not provide a sufficient explanation of personality, and there is more to personality than traits. Personality traits must be inferred through measurement, since they are hypothetical constructs that cannot be observed directly, although some trait psychologists, such as Hans Eysenck and Gordon Allport, view traits as “neuropsychic entities.” Support for this view has been provided by neuroscience findings that have suggested a genetic link to the major dimensions of personality traits; however, those findings are still preliminary, with half of the variance, at the most, being attributed to genes, and the environment or the interaction of traits with the environment accounting for the rest. Research into polygenetic influence on personality traits is emerging in light of several null replications of earlier single-gene studies.

Strictly speaking, a personality rating scale is a subset of items that all describe the same personality characteristic, variable, or trait along a continuum, with multiple categories that are assigned a number and can therefore yield a score. For example, in assessing the extent to which the item “cautiousness” describes a person, Likert scales may be used, such as “very much like the person”; “somewhat like the person”; “uncertain”; “somewhat unlike the person”; “not at all like the person.” The rater makes an evaluative judgment by choosing which category along the continuum most accurately depicts the person who is being rated (the ratee). Like physical and mental attributes, traits vary in the population in a continuous and normal distribution (bell-shaped), with most falling within the middle range and fewer lying within the extremes.

Because traits are pervasive across situations, scores on trait measures should be relatively consistent across time, and ratings on scale items measuring the same underlying trait should be in agreement, as should ratings on items assessing different aspects of the same trait, such as test retest and internal consistency reliability, respectively; both represent criteria for scientific soundness of scales set forth by psychometricians.

Personality can be described by rating oneself (self-ratings). Also, persons can be described based on the impressions that they make on others who observe them. Thus, informants rate another person based on their observations and perceptions of the individual that they are assessing. These are referred to as "observer ratings" and can be made by a peer, supervisor, teacher, or counselor. Research has also shown consistency between the different methods of assessing personality, providing support for the scale’s reliability validity (that is, that the scale is measuring the construct that it is intended to measure, an additional criterion set forth by psychometricians to evaluate the scientific soundness of personality assessment measures).

Lewis Aiken noted in 1997 that personality rating scales are included in inventories and may be a part of questionnaires. Both inventories and questionnaires are sometimes commercially labeled as “scales.” Regardless of what form they take and how they are labeled, personality rating scales are widely used in both research and applied settings. Various subdisciplines of psychology—such as personality, social, developmental, educational, school, industrial, clinical, and forensic psychology—rely extensively on rating scales for the scientific assessment of personality for research purposes. They may be used in applied settings, such as medical and health care, to assess behavioral risk factors; mental health treatment, to measure psychopathology; colleges, to assist with vocational guidance; business and industry, to aid in personnel selection; the armed services, to aid in selection of those who are the most fit; and criminal justice, in the area of profiling.

Origin, Development, and History

The foundation for the family of paper-and-pencil methods (of which personality rating scales are members) to obtain scientific information about people, products, or events evolved out of work in a variety of disciplines. Pierre-Simon Laplace and Carl Friedrich Gauss conducted seminal research on probability theory—the bell-shaped, normal distribution now termed "Gaussian distribution"—in the early eighteenth century, which made it possible to infer logically the characteristics of populations (whether physical, mental, or personality traits) from the analysis of sample data. Adolphe Quetelet extended Laplace’s and Gauss’s work to biological and social data, which marked the beginnings of vital statistics, that is, data pertaining to human life. Gustav Fechner’s work in the nineteenth century in the area of psychophysics (subjective mental events) and the mathematical measurement of physical stimuli that gave rise to them produced Fechner’s law, that sensation increases with the logarithmic value of the stimulus. Sir Francis Galton worked in the 1880s with Karl Pearson in the area of statistical methods and also contributed pioneering methodologies in the area of individual differences. The Woodworth Personal Data Sheet, the first formal, self-report, multi-item scale personality inventory that assessed psychoneurotic tendencies, was constructed by R. S. Woodworth in 1918 for purposes of weeding out unfit military personnel in World War I. Other early personality measures constructed in the first half of the twentieth century include the attitude inventories of L. L. Thurstone and his colleagues, the Strong’s Vocational Interest Blank for Men, the Vernon and Allport Study of Values, the Bernreuter Personality Inventory, and the Minnesota Multiphasic Personality Inventory.

Current Status and Applications

Since 1960, a number of scales have been created to measure traits and to organize them into a coherent structure or taxonomy. This effort was significantly advanced by the successful use of factor analysis in the development of a taxonomy of mental abilities and the use of high-speed computers in psychological research. This increased the efficiency and precision of factor analysis in grouping correlated items into factored scales that purportedly map the personality sphere.

Some have regarded trait systems as the theoretical model and scales and inventories as their application, although not all scales and inventories represent explicit theories. Raymond B. Cattell used an empirical approach in which he defined the “personality sphere” with words in the language that are used by observers to describe behavior. Others, such as J. Paul Guilford and Hans Eysenck, have been influenced by existing theory (Carl Jung’s psychological types) and their own factoring of existing items and scales in addition to the research of others. On the other hand, Andrew Comrey and Douglas Jackson have been less concerned with covering the total domain of personality. Instead, psychometric soundness has been the priority in their development of explicit procedures for item selection, scale construction, and validation in defining sets of primary trait scales that best define those areas of the domain that had been well researched.

These approaches have all yielded different numbers of factored scales: Cattell's has sixteen; Eysenck's, three; Guilford's, thirteen; and Comrey's, eight. A review and critical analysis by Saul Sells and Debra Murphy of the dimensions of personality represented by these factored scales pointed out considerable difference across systems, yet extensive overlap and common content. The differences are partially due to whether lower-order (narrower) or higher-order (more general) factors have been the focus, with Eysenck emphasizing the higher-order ones (that is, the superfactors of psychoticism, extroversion, and neuroticism) and Cattell, Guilford, and Comrey focusing on lower-order or primary factors. Some psychologists hold that separate scales based on the lower-order primary factors have greater richness and predict better than those with two scores that compress all of the information. Further, the utility of the levels may depend on the situation; for instance, for Eysenck’s psychophysiological experiments, the higher-order factors may be better and the primaries may be more suitable in the fields of clinical and personnel.

In attempts to resolve the discrepancies, there have been several investigations of congruences among the personality factor-trait systems, including one by Sells, Robert G. Demaree, and D. P. Will (the most ambitious project ever attempted in personality questionnaire research at that time) in which they administered Cattell’s and Guilford’s items to a sample of 2,500 individuals and factored them in the same analysis. The findings yielded eighteen factors, with five being common to both the Cattell and Guilford systems (emotional stability, social extroversion, conscientiousness, relaxed composure versus suspicious excitability, and general activity). With the inclusion of agreeableness, all but one (openness to experience) of the five broad bipolar dimensions that P. T. Costa Jr. and R. R. McCrae have deemed the Big Five were reflected in their results.

The Big Five are posited as accounting for most of the personality-attributable variation in human behavior, although at least thirteen other non-Big Five factors (six attributable to Guilford, five attributable to Cattell, and two attributable to neither) were extracted in the Sells, Demaree, and Will analysis. The two factors attributable to neither contained items that were similar in content but were dissimilar in terms of the source factors from which they were drawn. Expansion of the Big Five to include traits that are represented by combinations of pairs of factors has been suggested by some.

The research of Jackson and colleagues has yielded results that suggest that there are many dimensions of behavior beyond the Big Five and, further, that the narrower facets, thought to be subsumed under the broader Big Five factors, may provide more accurate behavioral prediction than the Big Five alone.

Further, the review by Sells and Murphy found that “two factors with labels similar to those emphasized by Eysenck, neuroticism and extroversion-introversion, were addressed across all five of the systems (Cattell, Eysenck, Jackson, Comrey, and Sells, Demaree, and Wills), although not exactly in the same manner or in the same terms.” They additionally point out that factor level can be “an artifact of the composition of the variables in the matrix” and that bandwidth of the various factors produced may be more important—that is, the extent to which the factors are very broad (made up of dissimilar items) or very narrow (made up of tautological items) may be more important than factor level in understanding the relationships among these factors.

In the light of the fact that personality factors and their measures are human constructs, research to determine the predictive validity of scales that already exist and those that are developed continues to be critical in assessing their theoretical and practical utility. Furthermore, debates on whether broad bandwidth factors or narrower band components have more predictive and explanatory power can best be served by examining each one’s specific situational utility within a hierarchical framework of the domain that encourages choice of assessment level.

Bibliography

Arroyo, Daniela, and Elias Delgadillo. Encyclopedia of Personality Research. Hauppauge: Nova Science, 2012. Print.

Butcher, James, and Julia Perry. Personality Assessment in Treatment Planning: Use of the MMPI-2 and BTPI. New York: Oxford UP, 2008. Print.

Ewen, Robert B. "The Trait Perspective." An Introduction to Theories of Personality. New York: Psychology, 2010. 239–86. Print.

Fiske, Susan T., and Patrick E. Shrout. Personality Research, Methods, and Theory.. New York: Taylor, 2014. Print.

Jackson, Marc-Antoine, and Evan F. Morris. Psychology of Personality. Hauppauge: Nova Science, 2012. Print.

John, Oliver P., Laura P. Naumann, and Christopher J. Soto. “Paradigm Shift to the Integrative Big Five Factor Taxonomy: History, Measurement, and Conceptual Issues.” Handbook of Personality Theory and Research. Ed. John, Richard W. Robins, and Lawrence A. Pervin. 3rd ed. New York: Guilford, 2008. 114–58. Print.

Kraus, Michael W. "Do Genes Influence Personality?" Psychology Today. Sussex, 11 July 2013. Web. 1 July 2014.

Paunonen, S. V., and M. C. Ashton. “Big Five Factors and Facets and the Prediction of Behavior.” Journal of Personality and Social Psychology 81. 3 (2001): 524–39. Print.

Paunonen, S. V., and D. N. Jackson. “What Is Beyond the Big Five? Plenty!” Journal of Personality 68 (2000): 821–35. Print.

Pope, Kenneth S., James Neal Butcher, and Joyce Seelen. The MMPI, MMPI-2, and MMP1-A in Court: A Practical Guide for Expert Witnesses and Attorneys. 3rd ed. Washington: Amer. Psychological Assn., 2006. Print.

Saucier, G., and L. R. Goldberg. “Evidence for the Big Five in Analyses of Familiar English Personality Adjectives.” European Journal of Personality 10 (1996): 61–77. Print.

Sells, S. B., and Debra Murphy. “Factor Theories of Personality.” Personality and the Behavior Disorders. Ed. N. S. Endler and J. M. Hunt. New York: Wiley, 1984. Print.