Who Am I? Well, It Depends: How Frame-of-Reference Imposes Context In Non-Contextualized Personality Inventories

摘要

The frame-of-reference (FOR) effect refers to the finding that validities for personality measures can be improved by asking respondents to consider how they behave in a particular context (e.g., u22at worku22). Recently, Lievens, De Corte, and Schollaert (2008) demonstrated that a FOR serves to reduce within-person inconsistencies in responding, which then improves the reliability and validity of personality measures. Despite this important step forward in FOR research, Lievens et al. note that there is still very little known with regard to how respondents complete non-contextualized personality inventories (i.e., inventories where no FOR is provided). The present studies sought to fill this significant gap in the literature by addressing the question: Do people think of themselves in particular situations or contexts when responding to non-contextualized personality inventories and, if so, what are these contexts? In addition, does the use of context vary by the personality dimension being studied? Two studies were conducted in order to fully address these Research Questions. The first of these studies was a qualitative study which examined the number and types of contexts spontaneously generated by test-takers for non-contextualized personality items. Twenty-eight interviews were conducted with college students who held a variety of life roles (e.g., student, employee, parent, spouse). Interview data demonstrated that participants considered themselves in general, at school, at work, with friends, with family, at home, and in other more specific situations (e.g., driving a car) when responding to non-contextualized inventories. Data for Study 2 were collected from 463 college students using a self-report methodology that asked participants to indicate which FORs they were using in responding to the same non-contextualized inventory used in Study 1. Results indicated significant differences in FOR endorsement across factors, such that participants endorsed the highest number of FORs for agreeableness items and the lowest number of FORs for openness to experience items. In addition, there were significant differences in the use of FORs within factors such that, for example, the u22With Familyu22 FOR was used most frequently for agreeableness but the u22At Schoolu22 FOR was used most frequently for openness to experience. Finally, results of Study 2 indicated that while the using more FORs in responding may increase error variances, it does not have a substantial impact on the factor structure of the Big 5. The present studies contribute to the literature by being the first to examine the role that situations play in responding to a non-contextualized inventory, and they do so using both qualitative and quantitative methods. In addition, the present studies represent a person-centric approach to the study of I/O psychology in that they focus on the individual experience as the basis for research.