The current thesis presents five studies which consider the structure of normal range human personality as measured by psychometric inventories. The primary aim of the thesis was to develop a hierarchical model of the structure of human personality, which not only encapsulated the content of extant personality inventories, but extended the extant models to better represent the accumulated knowledge in the field of personality research. To do so, the thesis reports on the largest item level exploratory and confirmatory factor analytic investigation of the structure of personality to date. In Study 1, the utility of confirmatory factor analysis (CFA) for the investigation of personality structure was established by conducting dimensionality, structural and invariance analysis of the 16 Personality Factor Questionnaire, Version 5, using the US Standardization sample. The results clearly demonstrated that confirmatory models of personality inventories can display model fit, and that structures can be identified which are invariant across samples. In Study 2, 1,772 items from the Eugene-Springfield community sample were used to locate 78 unidimensional narrow personality facets, which were both substantively consistent and displayed strong psychometric properties. These 78 facets were then used to identify a second order 11 factor global structure to human personality, the 11+ Model. The 11+ Model was robust over different rotational criteria, and displayed excellent theoretical coverage of the personality sphere. However, confirmatory factor analyses of the 11 factor solution failed to demonstrate model fit. As a result, three further studies were conducted to explore the possible reasons for model misfit. In Study 3, the model fit of the new exploratory 11 factor solution was considered against model fit for extant structures and inventories. The results showed that across modelling strategy, the fit of newly identified structure was comparable to the extant inventories. In Study 4, the claim the model misfit is caused by the inherent complexity in personality was considered through an application of exploratory structural equation modelling (ESEM). ESEM has been argued to more adequately represent such complexity. Results suggested that the ESEM framework fails to offer additional information to help resolve the substantive problems in personality research. Finally, in Study 5, the sample size required to reliably estimate single factor confirmatory models, and second order structural models, was investigated using Monte Carlo simulations. The results suggest that samples of over 2000 are required to reliably estimate second order structural parameters in confirmatory factor analytic models. This finding suggests the samples used in much published research to investigate confirmatory structural models of personality inventories may be deficient.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:553446 |
Date | January 2012 |
Creators | Booth, Thomas |
Contributors | Irwing, Paul; Batey, Mark |
Publisher | University of Manchester |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Source | https://www.research.manchester.ac.uk/portal/en/theses/a-review-of-the-structure-of-normal-range-personality(5ed8da05-4af3-43f4-952c-38edd8c89596).html |
Page generated in 0.0021 seconds