Robins+et+al.++2001

To test the structural stability of the big five we

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: was less than 5% (i.e., we used a 95% confidence interval). Rank-order stability was assessed by the correlation between Week 1 and Year 4 personality scores. To test the structural stability of the Big Five, we used structural equation modeling to compare the fit of two models, one in which the intercorrelations among the Big Five were freely estimated at each time point and the other in which the intercorrelations among the Big Five were constrained to be equivalent across the two assessments. A significant difference in fit between these two models would indicate that there was significant change in the structural relations among the Big Five. To assess ipsative stability over time, we computed three indices of profile similarity. Cronbach and Gleser (1953) observed that individual profiles can vary in three ways: elevation (the average level of scores), scatter (the variability of scores), and shape (the patterning of scores, or relative salience of the Big Five within a profile). Cronbach and Gleser outlined three methods for quantifying these sources of variability. The first index, D2, quantifies the 3. The RC index is based on classical test theory and thus makes a number of assumptions that may or may not be valid (e.g., error variance is constant across participants and over time). To the extent that these assumptions do not hold in the present data, our estimates of how many individuals increased, decreased, and stayed the same may be biased. 626 Robins et al. squared differences between trait levels at two time points, summed across all five traits. D2 is sensitive to differences in elevation, scatter, and shape. The second index, D′2, quantifies the squared differences between profiles after each profile has been centered around its mean. D′2 is insensitive to differences in mean levels between profiles and only reflects differences in scatter and shape. The third index, D″2, quantifies the squared differences between profiles after each profile has been standa...
View Full Document

Ask a homework question - tutors are online