Use of computer dictation by students with learning disabilities

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)




Lawrence J. Lewandowski


Speech recognition, Computer dictation, Learning disabilities

Subject Categories

Education | Educational Psychology


Twelve students with learning disabilities from grades five through eight composed stories in three conditions: handwriting; dictation using IBM's VoiceType 3.0 voice recognition software; and a modified condition in which students dictated but were not required to correct computer recognition errors. For the entire group of 12 students, dictated pieces contained significantly fewer spelling errors than handwritten pieces, but did not differ in measures of length, fluency and quality. For the six slowest copiers on a paragraph copying task, dictated compositions were significantly longer and tended to be more fluent than handwritten pieces. For the six fastest copiers, dictated compositions were shorter and tended to be less fluent. When students were permitted to dictate continuously, leaving computer recognition errors to be corrected by the experimenters, the entire student group dictated compositions that were significantly longer and more fluent than handwritten compositions, with no diminution in quality. Limited accuracy in recognizing the spoken word--and the need for students to correct recognition errors--confined the length benefit of VoiceType 3.0 to students who wrote slowly. Using dictation, these students were able to compose more, in less time, and with fewer spelling errors than they could using handwriting, and with no loss in quality. Moreover, most students preferred dictation to handwriting, suggesting that computer dictation may make the task of composing more appealing to many struggling writers. The paragraph copying task that was used may be useful in selecting those students who are likely to derive the greatest benefit from computer dictation. A broader segment of the student population may benefit from computer dictation as the accuracy of voice recognition increases.