Ratings and Measures of Cleft Palate Speech

Abstract

Rating procedures are commonly employed in the diagnosis of commu— nication efficiency of speakers with cleft palates to make assessments of speech intelligibility and articulation. Justification for the continued use of these convenient procedures should depend upon the reliability and validity of the ratings obtained. Considerable evidence to support the use of ratings of articulation is available. Good intra-and inter-judgment agreement for ratings of artic-ulation in the speech of individuals with cleft palates has been reported (6, '7). Positive and significant relationships between ratings of overall articulatory performance and number of errors as identified by conven— tional articulation testing have also been reported (4, 6', 9). In general, less evidence is available to support the continued use of overall ratings of intelligibility. In simple terms, measures of intelligibil-ity may be obtained by comparing the words comprehended by listeners with words intended by the speaker. Such measures are practical references for describing competence in communication and for expressing the communicative significance of disordered speech. Unfortunately, however, the procedures necessary for this type of measurement are extremely costly in time since they require carefully controlled tape recordings, playback to multiple listeners and rather tedious procedures in analysis. This latter factor is largely responsible for the common practice of rating overall intelligibility rather than measuring intelligibility as previously described. This study was undertaken: 1) to evaluate differences between clinical ratings of overall intelligibility and measures of intelligibility, and 2) to describe a clinical format for the organization of multiple speech ratings and measures. Intelligibility and articulation are known to be significantly related; however, the two attributes are not the same. For this