Criterion Validity of Speech Intelligibility Rating-Scale Procedures for the Hearing-Impaired Population

Abstract

The criterion validity and reliability of two popular rating-scale procedures for the assessment of the contextual speech intelligibility of hearing-impaired individuals (The NTID Scales) were studied under clinically typical conditions of test administration and evaluation. The criterion measure was a write-down procedure based on subjects' readings of sentence lists chosen from The CID Everyday Sentence lists. Although the results revealed generally high overall validity and reliability coefficients for the two scales, a close examination of the distribution of estimation error over the full range of rating-scale values revealed gross violations of measurement prediction within the clinically most frequent midrange of speech intelligibility. The results indicate that the rating-scale procedure significantly compromises clinical and research classification of an individual's speech intelligibility and that write-down procedures may provide a viable and significantly more accurate alternative for speech intelligibility assessment.