Wright, Sarah F.; (1991) Reading and spelling problems in school children: Their identification and prognosis. Doctoral thesis (Ph.D), UCL (University College London).

Text Reading_and_spelling_problems_.pdf Download (16MB)

Abstract

This five year longitudinal study focuses on (a) a comparison at two ages of two multivariate techniques for defining dyslexia; (b) the cognitive and behavioural characteristics associated with the presence of a large discrepancy between ability and reading and spelling achievement; (c) the rate of development of reading and spelling over a five year period; and (d) the extent to which reading ability can be accurately predicted. Regression and cluster analyses were used to identify dyslexic groups in a sample of 416 school children followed longitudinally from age eight to age thirteen. Concordance between the two methods at age eight was high, 95% of the dyslexics identified by regression were identified by cluster analysis. At age thirteen concordance was lower between the two techniques, with 67% of those identified by regression also identified by cluster analysis. Using each technique, three separate groups of dyslexics were identified. A stable group who were identified as dyslexic at both ages, a transient group who were identified as dyslexic only at age eight and a 'late developing' group who were identified as dyslexic at age thirteen but not at age eight. Six factors were derived from a cognitive and neuropsychological battery. These were interpreted as measuring: Visual and auditory memory and motor skills, Handedness, Spatial ability. Sequencing ability, Memory and Left right discrimination. Performance on three of these factors: Sequencing, Left-right discrimination and Visual-perceptual memory, differentiated the dyslexics and controls and contributed significantly to both the concurrent and future prediction of reading achievement. Results of a two factor (dyslexic vs non-dyslexic by Time) multivariate analysis of variance indicated that the reading performance of dyslexics, defined by both techniques, improved at a faster rate than that of controls over the five year period. Dyslexic readers, however, did not approach the levels of performance of the non dyslexics at either age. For measures of spelling and Verbal IQ, the rate of development was found to be similar for both groups. Prediction of later reading achievement was found to be as accurate as current prediction. Stepwise multiple regression equations indicated that at age eight a subset of cognitive variables and age accounted for 45%, 39%, 41% and 53%, respectively, of the variance in Schonell Reading, Schonell Spelling, Neale Accuracy and Comprehension. At age thirteen, the same variables, excluding age and including a measure of attention, accounted for 44% of the variance in Schonell Reading and 44% in Schonell Spelling. The best predictor of later reading was prose reading at age eight. When this measure was included in the analysis, the other cognitive variables contributed only small amounts of additional variance.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item