Robins+et+al.++2001

188 p 05 based on the levenes test for the equality of

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: personality or demographic dimensions, except that the variance for SAT scores was slightly lower among participants than nonparticipants (SD = 171 vs. 188, p = .05), based on the Levene’s test for the equality of variances. In light of these differences, the participants in the study may not represent a completely random, unbiased sample of the original study participants. Nonetheless, on most variables of interest, there were no differences between participants and nonparticipants, and our subsequent analyses are unlikely to be seriously biased by nonrandom attrition. 2. These response scale anchors are different from those typically used for the NEO-FFI (Strongly disagree, Strongly agree). This difference should be taken into consideration when making comparisons to norms from other samples. Personality Change 625 sample of 107 undergraduate students at one point in time and then 2 weeks later. The Big Five scales differed only minimally in their test-retest reliabilities: .86 (Extraversion), .86 (Agreeableness), .90 (Conscientiousness), .89 (Neuroticism), and .88 (Openness). Measuring Personality Change Mean-level change was assessed by comparing mean personality scores at Time 1 (Week 1) with mean scores at Time 2 (Year 4) using dependent samples t tests. To determine the degree to which the mean-level changes held for each participant in our study, we also assessed individual-level personality change. We classified people as having decreased, increased, or stayed the same on each dimension, based on the Reliable Change index (RC; Christensen & Mendoza, 1986). The RC index quantifies the probability of observing a difference score equal to or greater than the one observed, assuming that no change has occurred. The RC index explicitly accounts for the unreliability of measurement, thereby providing a useful method for separating true personality change from change due to measurement error (Jacobson & Truax, 1991).3 We classified an individual as having increased or decreased on each dimension if the probability associated with his or her RC score...
View Full Document

This note was uploaded on 12/02/2010 for the course PSYCH PSY BEH P2 taught by Professor Susanturkcharles during the Fall '10 term at UC Irvine.

Ask a homework question - tutors are online