Date of Award


Publication Type

Doctoral Thesis

Degree Name





Education, Educational Psychology.




At this point, the research efforts of many investigators of learning disabled children have most often included only those children who could be considered learning disabled after excluding all other possible reasons for their academic difficulties. These studies typically exclude all children who are considered low functioning (i.e., low-IQ), emotionally disturbed, environmentally deprived, brain damaged, and hearing or visually impaired. The purpose of the present study was to investigate the subtypal composition of various samples of learning disabled (LD) children selected on the basis of WISC FSIQ level. The issue of the adequacy of the traditional exclusionary definition of LD children was investigated in the present study. Two samples of LD children were selected from a population of 4800 children referred for neuropsychological assessment. Sample A (n = 1200) contained children usually excluded from LD studies (e.g., emotionally disturbed). This sample contained children who exhibited a relatively wide range of IQ values (i.e., WISC FSIQ 60-120). Sample A was initially subtyped on academic measures (i.e., WRAT subtest scores) in an attempt to provide statistically derived classifications of learning disability. This initial analysis was referred to as Phase I. The LD children in these Phase I subtypes were then cluster analyzed on 12 measures of neuropsychological abilities in order to further delineate their adaptive strengths and weaknesses. These analyses were referred to as Phase II. In this set of analyses Lorr's (1983) procedure for clustering first for profile shape and then for profile elevation was employed. A second set of samples were constructed on the basis of WISC FSIQ ranges from Sample B (n = 882). With the exception of low-IQ children all other types of unusual children were excluded from Sample B (WISC FSIQ 70-110). Four subsamples corresponding to WISC FSIQ ranges 70-80, 81-90, 91-100, and 101-110 were selected from Sample B and cluster analyzed with respect to 12 measures of neuropsychological abilities. Each subsample (n = 70) was cluster analyzed to determine if differences in subtype number and subtype structure would emerge as a result of WISC FSIQ level. These analyses were referred to as Phase III in the present research. Four WRAT subtypes emerged from the cluster analyses of Sample A (n = 1200) in Phase I. WRAT subtypes #1, #2, and #4 were considered very heterogeneous with respect to the 'types' of children that they contained. The expectations for the Phase I analyses were largely unsupported. It was the case that the Phase I analyses did not yield strong evidence either for or against the utilization of an exclusionary LD definition. Nine subtypes emerged from the Phase II analyses. Five of these subtypes contained relatively few low-IQ children. However, only one of these subtypes contained a majority of these low functioning children. Sixteen subtypes emerged from the cluster analyses of the Sample B subsamples. At least two profile shapes were found which occurred within each of these subsamples regardless of WISC FSIQ range.Dept. of Psychology. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis1990 .M385. Source: Dissertation Abstracts International, Volume: 52-11, Section: A, page: 3866. Thesis (Ph.D.)--University of Windsor (Canada), 1990.