Reanalyzing the 16pf-5 second order structure: Exploratory versus confirmatory factorial analysis |
| |
Authors: | Anton Aluja Angel Blanch Luis F García |
| |
Institution: | (1) National Institute on Aging,NIH,DHHS, Baltimore, USA;(2) Johns Hopkins Bloomberg School of Public Health, Baltimore, USA |
| |
Abstract: | In the present study, the second order structures of the 16PF-5 for 3, 4, 5 and 6 factors are analysed in a sample of 636
undergraduate students and their friends and relatives, with a mean age of 25.09 years (sd: 9.20). A two-stage analysis with
Exploratory and Confirmatory Factor Analysis (EFA and CFA) was performed assigning subjects either to a calibration sample
(314 subjects), or to a validation sample (322 subjects). First, EFAs were conducted for each of the 4-factor structures with
the calibration sample, and second, CFAs were specified and assessed also for the 4-factor different structures. Four different
situations were tested, simple structure, secondary loadings >.30, modification indices between covariance error terms >20,
and modification indices between covariance error terms >15. Results show that the 16PF-5 simple structure yields poor fit
indices throughout all models, although when incorporating secondary loadings and correlating the variances of the error terms
with modification indices >15, the 5-factor model appears to be the best representation of observed data. The usefulness of
CFA in the study of personality structural questionnaires is discussed in the light of these results. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|