Individual Submission Summary
Share...

Direct link:

Age Differences in Big Five Characteristics across Varying Test Lengths

Fri, October 5, 10:45am to 12:15pm, Doubletree Hilton, Room: Redrock

Abstract

The field of personality has dedicated considerable attention to describing age differences in the Big Five personality traits. Although there is a rough consensus on general patterns and properties of age differences in personality (Roberts & Mroczek, 2008), there is still considerable variability in patterns found even among comparable samples (Donnellan & Lucas, 2008; Lucas & Donnellan, 2009; Soto, John, Gosling, & Potter, 2010; Srivastava, John, Gosling, & Potter, 2003). Part of this variability might arise from true differences between samples; other variability might arise from the type of Big Five measure administered. In the current study, we examine how one particular methodological feature, test length, might affect how age differences in personality are manifest. Using secondary data from a large study of online social networking behavior (N = 1,222,208; 60.4% female), we examined age differences in personality from age 13 to 112 across varying lengths of the same Big Five questionnaire, the International Personality Item Pool (IPIP; Goldberg, 1999; Goldberg et al., 2006). We had a large number of participants for each age group complete measures, albeit the sample skewed young (Mage = 26.18, SD = 9.39; ages 13-19:N = 197,340; 20-29:744,821; 30-39:176,085; 40-49:69,430; 50-59:23,422; 60-69:5,381; 70-79:1,312; 80-89:872; 90-99:1,233; 100+:2,312). Participants selected to take one of nine different versions of the IPIP—20, 30, 40, 50, 60, 70, 80, 90, or 100 items. Overall age differences were consistent with previous research on personality development. Specifically, agreeableness and conscientiousness were higher among older adults and lower among younger adults (βs ranged from |.03| to |.29|). Extraversion, neuroticism, and openness to experience were higher among younger adults and lower among older adults (βs ranged from |.01| to |.22|). Conscientious individuals were more likely to choose to complete the longer measure; measure length was unrelated to mean differences in the other four traits. Importantly, measure length moderated many of the associations between age and personality (βs ranged from |.02| to |.06|). Decomposing these interactions revealed that age differences across measure length were largely consistent among younger adults but diverged among middle aged and older adults. The findings will be discussed in the context of how much information is acquired from measures of different lengths. Specifically, we will include a discussion of measurement invariance both across age and inventories to provide practical recommendations for personality researchers interested in measuring the development of personality across life.

Authors