The Guilford-Zimmerman Temperament Survey (GZTS) was developed in 1948 by J. P. Guilford and W. S. Zimmerman for measuring normal personality development and temperament. The reason for developing personality trait measuring instruments in the field of psychology was the simplification of tests used to measure personality traits. Early tests usually focused on theories relevant to particular traits, which resulted in the creation of several hundred different tests, each measuring only one trait (McCrae & Costa Jr., 2012). Though many tests had good psychometric properties, there was no unified personality test, so the administration and interpretation of different test was disorganized and inefficient for both applied psychology and research. That is why Guilford and Zimmerman aimed to create a single test consisting of multiple inventories of traits, which would simplify personality testing.
Like most other early personality measurement instruments, the GZTS was based on Jung’s theory of psychological types, which became a popular theory upon which many researchers built personality trait scales (McCrae & Costa Jr., 2012). In their early research, J. P. Guilford and R. B. Guilford applied factorial analysis to personality research to analyze the measures of Introversion-Extroversion, one of the most popular aspects of Jung’s theory (McCrae & Costa Jr, 2012). They found five different traits associated with the 36 items reported by the participants, which resulted in the preparation of items that would be used in scales for measuring those traits. Until 1948, J. P. Guilford continued performing factorial analyses of personality, which led to the development of the GZTS as an inventory of 10 traits.
GZTS was subsequently used in various longitudinal researches aimed at investigating age changes to determine its validity and retest stability. One of the most notable studies that used the GZTS was the Baltimore Longitudinal Study of Aging (BLSA) by Terracciano, McCrae, and Costa (2006), which found that the test has high internal consistency, structural stability, and test-retest stability. The diversity of the sample in the BLSA study was poor because women were not included in the study and non-White males represented only a fraction of the sample, but McCrae and Costa Jr. (2012) argue that other studies with diverse samples managed to replicate the results from the BLSA.
Despite the good psychometric properties of the GZTS, several other personality tests were created using different theories and models. Consequently, the creators competed for the dominance of systems they had developed, so research papers could not be replicated and compared unless they used the same tests and models. For example, a study of age changes using the GZTS could not be replicated by scholars using the Murray’s system of needs (McCrae & Costa Jr, 2012). Today, the five-factor model is been widely accepted as the standardized taxonomy of personality traits in the academic community, so literature reviews and research in personality psychology mainly rely on that model along with the Revised NEO Personality Inventory and Big Five Inventory (McCrae & Costa Jr., 2012).
While the five-factor model is currently dominant in personality research, GZTS is still used by personality psychologists, counselors, and organizational psychologists for non-clinical assessments of individuals over 16 years of age. The targeted populations can include high-school students, college students, and adults. The test is currently published by Pearson Clinical and is available for purchase through their website (http://www.pearsonclinical.com/). The GZTS Handbook is priced at $49 while the package of 25 GZTS Test Booklets is priced at $23. Although the test can be scored manually with the GZTS Handbook and the Test Booklets, Pearson clinical offers the GZTS Q Local Interpretive Reports software for $25.75 and Mail-in Interpretative Reports for $29.00 per report.
GZTS has an inventory of 10 traits. Each of the 10 scales measures one of the following personality traits: (1) general activity, (2) emotional stability, (3) restraint, (4) objectivity, (5) personal relations, (6) masculinity, (7) ascendance, (8) friendliness, (9) sociability, and (10) thoughtfulness. Each scale has 30 items that are presented as self-descriptive statements. The response is recorded on a 3-point scale (1 = yes, 2 = no, 3 = not sure).
Proper administration for the GZTS must adhere to the common principles of personality trait test administration. For example, the administrator should be in control of the physical environment to ensure adequate lighting, temperature, and ventilation in the room while reducing noise and distractions. Some other factors that could affect the test outcomes must also be considered depending on the sample and setting used for the test administration. For example, the GZTS can be used in schools for adolescents over 16 years of age, so the scheduling of the test should not interfere with their other activities.
Another critical requirement for administering the GZTS test is obtaining informed consent from the participants. The participants must provide informed consent because the testing with the GZTS is not mandated by law. In some cases, testing is a regular part of school activities or a mandatory requirement for obtaining a position in an organization. In those cases informed consent is not required, but the test administrator must still provide accurate information regarding the purpose of the test, give test takers proper instructions, and ensure the participants are given an adequate time frame for completing the test. For the GZTS, the time it takes to complete the test is between 30 and 60 minutes.
The test administrator must also satisfy certain criteria to ensure proper administration procedures are met. For administrating the GZTS, a level C certification is required (Pearson Clinical, 2014). That means the test administrator must have a doctorate degree in psychology or a similar field that requires training in test administration procedures, scoring, and interpretations. Alternatively, the individual must have a license or certification to practice personality assessments or have an active membership with an organization relevant to the field of assessment (e.g. American Psychiatric Association). Those restrictions apply because improper administration of the test might produce inaccurate results.
The scoring can be done manually or with the GZST Interpretative Report, which computes the raw scores, C scores, and T scores for each. The interpretative report also provides commentary on the results. Manual scoring is done by adding up all the answers with values 1 or 2 while answers with the value 3 are not included in the calculation. However, the scoring may need to be reversed for certain scales. For example, to get the raw score on the emotional stability scale, it would be necessary to recode the responses after obtaining the initial sum of items answered with 1 or 2 by reversing the items with answers of 1 and 2, followed by adding up the new items with the value 1 to the existing score.
Once the raw scores are obtained, they need to be transformed into standardized scores. The interpretative report provides C scores and T scores for all scales as long as the amount of uncertain responses is acceptable. Even if those cases occur, the results can be computed manually with the assistance of the GZTS Profile Chart, which serves as a reference for translating GZTS raw scores into standardized scores. The GZTS is scored by a normality-based key, which means the norms for interpreting results are derived from a general adult sample. Therefore, once the standardized scores are obtained, it is possible to determine the percentile rank of a single individual and interpret their results by comparing it to the norm.
The GZTS has good psychometric properties. It has a high internal consistency with coefficients ranging from .75 to .87 (Terracciano et al., 2006). Longitudinal studies also tested the structural stability of the survey. It was found that the GZTS test-retest reliability coefficients range between .75 and .91 while the retest stability coefficients for a 24-year interval are between .61 and .71 (Terracciano et al., 2006).
The GZTS is mainly criticized for using 10 separate scales, even though it would be possible to summarize the items into only four broader scales without losing the test’s effectiveness (i.e. Introversion-Extraversion, Emotional Stability, Social Activity, and Paranoid Disposition). Nevertheless, the GZTS is often recommended as a reliable and stable personality assessment instrument because it generates accurate information (Terracciano et al., 2006).
Another limitation of the test is the potential of participant bias. Linden and Olson (1959) found that participants might provide fake answers on the Personal Relations scale to present themselves favorably. However, that limitation is common to all self-reported questionnaires and can be counteracted with various strategies, such as observing the timing of responses for online tests or warning the participants that there are detection methods in place for fake answers.
Finally, it is important to note that the test results are non-clinical. Therefore, they should never be used to determine a psychological disorder diagnosis. If the test administrator considers that results on certain scales indicate the presence of a pathological problem, further testing with tests developed for clinical testing (e.g. Minnesota Multiphasic Personality Inventory) to confirm the suspicion is required. Otherwise, the GZTS scale can be used only for non-clinical purposes, such as counseling, career planning, or performing organizational assessments.
Linden, J. D., & Olson, R. W. (1959). A comparative analysis of selected Guilford‐Zimmerman Temperament Survey scales with the Taylor Manifest Anxiety Scale. Journal of Clinical Psychology, 15(3), 295-298.
McCrae, R. R., & Costa Jr, P. T. (2012). Personality in adulthood: A five-factor theory perspective. New York, NY: Guilford Press.
Pearson Clinical. (2014). Qualifications policy. Retrieved from http://www.pearsonclinical.com/ psychology/qualifications.html
Terracciano, A., McCrae, R. R., & Costa, P. T. (2006). Longitudinal trajectories in Guilford–Zimmerman temperament survey data: Results from the Baltimore longitudinal study of aging. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 61(2), P108-P116.