Return to search

Exploratory and Confirmatory Factor Analysis of the Clinical Scales of the Luria-Nebraska Neuropsychological Test Battery, Form II

The factor structure of the Luria Nebraska Neuropsychological Battery (LNNB) Form II was examined. A principle components factor analysis was performed on a sample of 102 psychiatric and neurologic subjects. It was necessary to remove 45 items from the analysis due to perfect performance by most subjects. The results were orthogonally rotated to simple structure using a Varimax method of rotation, and then compared to previous LNNB Form I and Form II results. Thirty-three factors were generated in the Exploratory Factor Analysis (EFA) . There was a very high agreement with the factors from Form I. Only one new factor was identified that didn't have a comparable Form I factor, and this factor appears to have neurological support. The similarity of the factor solutions between the two forms supports the continued use of factors derived from Form I for the interpretation of Form II, and supports the underlying structure presupposed by Lurian constructs. The present study also tested the significance of the hypothesized factor structures through confirmatory factor analysis (CFA). No hypothesis about the underlying factor structure based on previous exploratory studies was supported. The CFA did suggest that the best factor solution to the LNNB Form II is one that (a) has correlated factors and (b) has items loading on more than one factor. The confirmatory results were interpreted as not supporting the current exploratory results, or the previous factor analytic results. Problems notwithstanding, researchers may be better directed to propose factor models for the LNNB that have correlated factors, and to work samples approaching the 10 to 1 recommended sample size for multivariate analysis. One conclusion that was drawn from the concurrence between the two Form II studies pertains to psychiatric populations used in both studies. It was necessary to exclude a large number of items in each study due to perfect performance by most of the subjects on those items. Most of the items removed were identical in both studies supporting the notion that a shortened version of the LNNB could be administered to psychiatric populations.

Identiferoai:union.ndltd.org:unt.edu/info:ark/67531/metadc330984
Date05 1900
CreatorsNagel, Jeffrey A.
ContributorsHarrell, Ernest H., Martin, Sander, 1939-, Hresko, Wayne P., Martin, Charles B., 1930-, Beyerlein, Michael Martin
PublisherUniversity of North Texas
Source SetsUniversity of North Texas
LanguageEnglish
Detected LanguageEnglish
TypeThesis or Dissertation
Formatvii, 189 leaves, Text
CoverageUnited States - Texas
RightsPublic, Nagel, Jeffrey A., Copyright, Copyright is held by the author, unless otherwise noted. All rights reserved.

Page generated in 0.0016 seconds