Nonword Repetition And Sentence Repetition In Preschool Children: A Comparison Of Speech And Language Screening Measures

Nonword repetition tasks and sentence repetition tasks have been shown to accurately differentiate children with speech and/or language disorders from their typically developing peers (Dollaghan & Campbell, 1998; Archibald & Gathercole, 2006). Additionally, they provide information on semantic, syntactical, and phonological processing as well as memory and acoustic perceptual skills (Vance, Stackhouse, & Wells, 2005). Despite their utility as a screening measure, their use in clinical practice has been minimal. Instead, practicing speech-language pathologists often choose to use publisher-provided screening measures or their own developed measures, rather than using research-based screening methods that have recognized differential diagnosis ability (Nash, Leavett, & Childs, 2011). The purpose of the present study is to compare the current speech and language screening measures used by student clinicians in the UND Speech, Language, and Hearing Clinic to the Dollaghan and Campbell's (1998) nonword repetition task (NRT) and the sentence repetition task (SRT) taken from the Clinical Evaluation of Language Fundamentals - Preschool Second Edition (CELF-P2) (Semel, Wiig, & Secord, 2004). In order to determine the clinical usefulness of the SRT and NRT as a speech and language screener, potential cut-off points to distinguish typically developing (TD) children and those who may have language impairment (LI) and/or speech sound disorder (SSD) were also considered. Forty-nine English-speaking, monolingual, preschool-aged children (ages 3;1 to 5;9 years) were recruited for this study. Results indicated a limited correlation was found between UND's measures and the SRT and the NRT. In addition, a strong relationship was found between the SRT CELF-P2 scaled scores and the percent consonant correct (PCCs) and percent phoneme correct (PPCs) scores of the SRT and NRT. Overall, a greater number of participants performed more poorly on the SRT and NRT than on UND's screening measures. These results indicate that the SRT and NRT may be detecting speech and/or language disorders that UND's measures fail to detect.