This study has two parts: Part I discusses the limitations of difference scores and exploratory factor analysis for representing speed of information-processing stages in the context of a reanalysis of a study by Vernon (1983). Vernon interpreted the differences between objectively measured reaction times on various simple cognitive tasks as components of speed of information processing. Correlations were calculated among these differences and subjected to exploratory factor analysis. The factors obtained from this analysis were interpreted by Vernon in terms of short-term and long-term memory processing constructs. The use of difference scores, however, implies an additive model allowance for random error, which correlations between these differences. that does not make leads to spurious The application of exploratory factor analysis to among these differences compounds uncover latent variables the problem because it admits many alternative interpretations which cannot be tested against one another for goodness-of-fit to the data. Confirmatory factor analysis addresses these problems. This thesis demonstrates that the correlations between the difference scores can be accounted for in terms of factors obtained from factor analysis of the original reaction time data. These factors lead to an alternative interpretation of the results which is contrasted with Vernon's interpretation.
Part II of this study illustrates the use of confirmatory factor analysis with this kind of data. An attempt to test the assumptions of Vernon's difference score model with confirmatory factor analysis did not succeed because the implied model was too constrained for the statistical program we were using; consequently, the program could not find a starting solution. In order to demonstrate how confirmatory factor analysis can be used to test models of speed of cognitive processing, Part II partially replicates a study by Lansman, Donaldson, Hunt, & Yantis (1982). This research analyzed a simple cognitive reaction time task that was examined in detail by Vernon. Donaldson (1983) used the Lansman et al. data to compare difference scores and part correlational techniques with a general approach based on analysis of covariance structures to demonstrate how the components of cognitive processes can be explicated using confirmatory factor analysis.
| Identifer | oai:union.ndltd.org:pdx.edu/oai:pdxscholar.library.pdx.edu:open_access_etds-4842 |
| Date | 01 January 1988 |
| Creators | Uhland, Gary A. |
| Publisher | PDXScholar |
| Source Sets | Portland State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Dissertations and Theses |
Page generated in 0.0019 seconds