Conroy, Steven (2011) Examining the Utility of a Multivariate Analytical System to Determine Clinically Significant MMPI-2 Profiles. Coursework Masters thesis, University of Southern Queensland. (Unpublished)
Abstract
Previously, research examining both cognitive and psychosocial inventories has utilised prototypical cognitive profiles and codetypes, respectively, in order to identify the extent of an individual’s cognitive impairment or psychopathology. More recently, contemporary research has challenged these previously entrenched assumptions, with examination of multivariate statistical analyses in order to determine more robust and empirically sound methods of allocation. Using a battery of neuropsychological tests, Dawes (2004) utilised cluster analysis, the outlier statistic Mahalanobis Distance (MD), and discriminant functions analysis (DFA) to classify and allocate cases with a variety of neurological and psychiatric symptoms into differing groups based on their cognitive strengths and weaknesses. Dawes identified five separate cognitive profiles and was able to identify profile membership with accuracy in excess of 88%. Little is currently known, however, regarding the efficacy and accuracy of Dawes’ method of analysis when applied to a psychosocial inventory such as the MMPI-2. Accordingly the goal of the present dissertation was the production of psychological profiles, suggest their meaning, and to demonstrate an empirically sound statistical method to allocate individual MMPI-2 cases from a large medico-legal sample, to psychologically meaningful cluster profiles. The first study utilised factor analysis to reduce the complexity of 75 dependent variables from each of the 3105 cases to a more manageable number for clustering. Results identified a ten-component final solution that provided the least number of factors and accounted for 74.60% of the total variance. Using an Agglomerative, hierarchical average linkage (between groups) clustering algorithm, study two analysed the ten-factor solution to determine the number of psychological profiles. Following three checks of appropriate cluster identification, a K-means cluster analysis was conducted utilising random aggregation centres with a prescribed five-cluster solution. The third study examined a means of allocating individual cases to derive clusters of psychopathology utilising empirically sound statistical analyses. This was achieved by the allocation metric MD in an attempt to correctly allocate cases to samples as a function of multivariate similarity. Matrix algebra was utilised to calculate the MDs and these were subjected to DFA. Results identified an accurate means of allocating a particular case to the appropriate cluster with an accuracy of correct classification in excess of 81%. Results from the current study lend support to recent assertions that there are no unique patterns of MMPI2 Basic scale scores, or codetypes, that characterise a particular diagnosis. A basic implication of the finding is that clinicians now have the means to allocate individual cases to pre-determined clusters of symptomatology using an empirically sound statistical methodology that reflects the complexity of human behaviour. This research ensures a means of psychological profiling that will not only inform, but also ensure that valuable, time intensive, and sometimes-costly treatment resources are allocated to those individuals that are most likely to benefit from them.
|
Statistics for this ePrint Item |
| Item Type: | Thesis (Non-Research) (Coursework Masters) |
|---|---|
| Item Status: | Live Archive |
| Additional Information: | Current UniSQ staff and students can request access to this thesis. Please email research.repository@unisq.edu.au with a subject line of SEAR thesis request and provide: Name of the thesis requested and Your name and UniSQ email address |
| Faculty/School / Institute/Centre: | Historic - Faculty of Sciences - Department of Psychology (Up to 30 Jun 2013) |
| Supervisors: | Graeme Senior |
| Qualification: | Doctor of Psychology (Clinical) |
| Date Deposited: | 16 Oct 2025 02:52 |
| Last Modified: | 16 Oct 2025 02:52 |
| Fields of Research (2008): | 17 Psychology and Cognitive Sciences > 1799 Other Psychology and Cognitive Sciences > 179999 Psychology and Cognitive Sciences not elsewhere classified |
| Fields of Research (2020): | 52 PSYCHOLOGY > 5299 Other psychology > 529999 Other psychology not elsewhere classified |
| URI: | https://sear.unisq.edu.au/id/eprint/52225 |
Actions (login required)
 |
Archive Repository Staff Only |
