Megan Hodge1, Carrie L Gotzke. 1. Department of Speech Pathology and Audiology, University of Alberta, Edmonton, Alberta, Canada. megan.hodge@ualberta.ca
Abstract
OBJECTIVE: This study describes a preliminary evaluation of the construct and concurrent validity of the Speech Intelligibility Probe for Children With Cleft Palate. DESIGN: The study used a prospective between-groups design with convenience samples. PARTICIPANTS: Participants (ages 39 to 82 months) included 5 children with cleft palate and 10 children with typical speech development and no history of craniofacial abnormalities. All children had age-appropriate language skills. INTERVENTIONS: Each child completed the Speech Intelligibility Probe for Children With Cleft Palate by imitating single words. Each child's word productions were recorded and played back to listeners who completed open-set and closed-set response tasks. Recorded utterances that represented a contiguous 100-word sample of each child's spontaneous speech also were played back to listeners for completion of an open-set word identification task. MAIN OUTCOME MEASURES: Measures reported include group means for (1) intelligibility scores for the open-set Speech Intelligibility Probe for Children With Cleft Palate and spontaneous speech sample conditions, and (2) percentage of phonetic contrasts correct and correct-distorted from the Speech Intelligibility Probe for Children With Cleft Palate closed-set response task. RESULTS: The group of children with cleft palate had significantly lower intelligibility scores, lower percentage of correct phonetic contrasts, and higher percentage of correct distorted items (construct validity). A strong positive correlation (r = .88, p < .01) was found between intelligibility scores from the Speech Intelligibility Probe for Children With Cleft Palate and the spontaneous sample (concurrent validity). CONCLUSIONS: The results provide preliminary support for the construct and concurrent validities of the Speech Intelligibility Probe for Children With Cleft Palate as a measure of children's speech intelligibility.
OBJECTIVE: This study describes a preliminary evaluation of the construct and concurrent validity of the Speech Intelligibility Probe for Children With Cleft Palate. DESIGN: The study used a prospective between-groups design with convenience samples. PARTICIPANTS: Participants (ages 39 to 82 months) included 5 children with cleft palate and 10 children with typical speech development and no history of craniofacial abnormalities. All children had age-appropriate language skills. INTERVENTIONS: Each child completed the Speech Intelligibility Probe for Children With Cleft Palate by imitating single words. Each child's word productions were recorded and played back to listeners who completed open-set and closed-set response tasks. Recorded utterances that represented a contiguous 100-word sample of each child's spontaneous speech also were played back to listeners for completion of an open-set word identification task. MAIN OUTCOME MEASURES: Measures reported include group means for (1) intelligibility scores for the open-set Speech Intelligibility Probe for Children With Cleft Palate and spontaneous speech sample conditions, and (2) percentage of phonetic contrasts correct and correct-distorted from the Speech Intelligibility Probe for Children With Cleft Palate closed-set response task. RESULTS: The group of children with cleft palate had significantly lower intelligibility scores, lower percentage of correct phonetic contrasts, and higher percentage of correct distorted items (construct validity). A strong positive correlation (r = .88, p < .01) was found between intelligibility scores from the Speech Intelligibility Probe for Children With Cleft Palate and the spontaneous sample (concurrent validity). CONCLUSIONS: The results provide preliminary support for the construct and concurrent validities of the Speech Intelligibility Probe for Children With Cleft Palate as a measure of children's speech intelligibility.
Authors: Gabriel J Cler; Talia Mittelman; Maia N Braden; Geralyn Harvey Woodnorth; Cara E Stepp Journal: J Speech Lang Hear Res Date: 2017-06-22 Impact factor: 2.297