A Test of Lip Reading Ability

Publisher Website
Google Scholar
Abstract
A list of 100 words was compiled from Thorndike''s list of most frequently used words by extracting every tenth word from the first 1000. To this was added a sentence test made up of 10 trial statements, 50 common expressions and 20 idiomatic sentences. The tests were first administered to a group of 200 subjects of normal hearing arranged so that the subjects had to lip read the material. Even the youngest child could score on the test. After arranging it in apparent order of difficulty the revised test was administered to 110 students in two schools for the deaf, all known to have at least third grade reading ability. The handicapped group showed a higher degree of proficiency in lip reading than even the best group of normals. Six paragraphs were then selected from the Gates Reading tests and added to the former test. This material did not contribute statistically to the test. The tests were then shortened by removing the most complicated and simplest material. The material was then photographed on black-and-white 16-mm. film for Sentence and Word Tests and Story Tests were added in technicolor. Each test situation was followed by a fifteen second period of black film to allow the individual to write what he thought he saw said. The test, as described, was then administered to 761 cases 8-21 yrs. from the 1st to the 12th grade with the age of onset of deafness ranging from birth through 15 yrs. For 702 cases, data reporting the cause were available: ''Acquired'' deafness 303 cases, ''congenital'' deafness 256, ''unknown'' deafness 135, and otosclerosis in 8 cases. General lip reading ability was reported by teachers of these hard of hearing children. The conclusions were as follows:[long dash] The skills of word, sentence and story recognition are interrelated. There is a great deal of overlapping among the various skills, yet, the combined skills do not represent a single unitary ability; word, sentence and story recognition should be measured separately for diagnostic purposes. Ability to lip read sentences can be predicted more reliably from ability to lip read stories than from ability to lip read words; ability to lip read words can be more reliably predicted from ability to lip read sentences than from ability to lip read stories; and ability to lip read stories can be more reliably predicted from ability to lip read sentences than from ability to lip read words. The internal reliability of each part of the test was shown to be high enough for practical purposes. The coeff. for the Sentence Test was 0.928, the coeff. for the . Word Test 0.797, and for the entire test 0.943. Reliable prediction of lip reading ability cannot be made from teachers'' judgment ratings. Lip reading ability cannot be predicted from reading level or school achievement, nor from chronological age, age of onset of deafness, or grade placement. The best criterion to be used as the basis for the standardization of a test of lip reading ability is the distribution of scores in percentile ranks. It was not possible to arrange the test items in any order of difficulty. The successes and errors made by the final group tested are distributed evenly throughout the whole test. The highest scores are obtained from a higher over-all response, poorer scores are mainly due to failure with one or another specific group of items.

This publication has 0 references indexed in Scilit: