'To be' or 'I am someone who tends to be': Does the wording of personality items matter?

The fundamental unit of data collection for self- and informant-report personality research is the item. In personality structure research, items have historically taken the form of single-word trait descriptors (e.g., “Calm”) while, in most personality assessment frameworks, they consist of phrases and sentences (e.g., “I am someone who tends to be calm”). This Registered Report will use a randomized, longitudinal design to evaluate the effects of differences in item wording on how people respond to personality measures. Specifically, we will examine the effect of four item formats—trait-descriptive adjectives presented alone (e.g., “Talkative”), with the linking verb “am” (e.g., “Am talkative”), with the additional verb “tend” (e.g., “Tend to be talkative”), and with the additional pronoun “someone” (e.g., “Am someone who tends to be talkative”)—on how people respond to an adjective-based measure of the Big Five (i.e., the Midlife Development Inventory; Lachman & Weaver, 1997). In terms of outcomes, we will consider item response distributions, participant response durations, and observed scale reliabilities, as well as the participants’ subjective experiences completing the survey. We will also test whether adding the pronoun “I” to any of these four formats influences the results. The findings from this work will have implications for ongoing research related to the assessment of personality, including the development of more detailed taxonomies of personality, the creation of more generalizable assessment models, and the harmonization of existing assessment models.