A comparison of frame-of-reference and frequency-based personality measurement. (April 2016)

• Record Type: Journal Article
• Authors: Robie, Chet; Risavy, Stephen D.
• Is Part Of: Personality and individual differences. Volume 92(2016)
• Page Start: 16
• Page End: 21
• Publication Date: 2016-04
• Subjects: Personality measurement -- Frame-of-reference -- Frequency-based estimation; Personality -- Periodicals; Individuality -- Periodicals; Individuality -- Periodicals; Personality Development -- Periodicals; Personnalité -- Périodiques; Individualité -- Périodiques; 155.205
• DOI: 10.1016/j.paid.2015.12.005
• Languages: English
• ISSNs: 0191-8869

Abstract

Abstract: Personality researchers and practitioners alike have historically been interested in improving the predictive validity of personality measures. Frame-of-reference and frequency-based estimation are two methods for improving personality prediction that have been studied in isolation. The present study examined the separate and combined effects of both of these methods on the predictive validity of a personality measure in a school setting ( N = 933 university students). The study used a 2 × 2 design to examine whether the relationships between two targeted personality traits (Conscientiousness and Extraversion) and relevant criteria (GPA and leadership ability) were affected by manipulating contextualization (noncontextualized versus contextualized) and response format (Likert-type versus frequency-based). The current study, in contrast to previous studies, did not find moderating effects for either method. We discuss possible reasons for our contrarian findings. Both methods have obvious (e.g., frequency-based measures take longer to complete and invoke more participant fatigue) and potential (e.g., contextualizing instruments is time-consuming and may affect psychometric properties) drawbacks that, in light of our findings, should give practitioners pause regarding the widespread implementation of either method for operational use. Highlights: Frame-of-reference (FOR) and frequency-based estimation (FBE) were examined. The two methods in isolation or in tandem did not increase predictive validity. Contextualization may not be consistently useful in increasing predictive validity. FBE was found to be mentally fatiguing and testing time was longer than Likert-type. Widespread operational use of either method may be unwise at present.