READING SKILLS OF SEVERELY LANGUAGE-IMPAIRED CHILDREN (DISORDERED, ACADEMIC, INTERMEDIATE).

The research literature has substantiated that many children diagnosed as having oral language impairments demonstrate reading problems at school-age. Few studies, however, have investigated the achievement of language-impaired children on individual reading skills, or the relationship between type of language deficit and type of reading disability. The purpose of this study was to investigate the performance of children with receptive language impairments (Receptives), children with expressive language impairments (Expressives), and children with no identified language or learning problems (Controls) on the reading skills of word attack, word recognition, vocabulary and comprehension. The subjects, ages 10-0 to 12-11, were 29 children enrolled in self-contained classes for the severely language impaired (SLI), and 37 normal children. The SLI subjects were categorized as Receptives (n = 24) or Expressives (n = 5) based on their performance on the Test of Language Development-Intermediate (TOLD-I). Reading scores were obtained for four subtests of the Woodcock Reading Mastery Tests-Revised (WRMT-R) and for two subtests of the Iowa Tests of Basic Skills (ITBS). Results of univariate analyses demonstrated Controls performed significantly (p < .05) better on all of the reading subtests than did either of the language-impaired groups. No difference was found between the Receptives and the Expressives on any subtest, possibly due to the small number of Expressive subjects. Because no differences were found between the language-impaired groups, they were combined to form a single group. Results of correlational analyses demonstrated correlations above .45 among all reading subtests for the Controls, indicating they measure similar abilities. For the SLI group, the subtests of the WRMT-R were highly correlated with each other (r < .65), but shared moderate correlations (.27 to .49) with the ITBS subtests. For the SLI group, the two ITBS subtests shared a low (.24) correlation. The Control group demonstrated reading achievement within normal limits, whereas the SLI group scored at least 2.4 years below the Controls on all subtests. For each group, profiles of subtest performance were flat, indicating that neither group demonstrates particular reading strengths or weaknesses