Psychometric properties of the mandarin clinical evaluation of language for preschooler's core scale

Purpose: To test the psychometric properties of a newly developed Mandarin Clinical Evaluation of Language for Preschooler's Core Scale (MCELP-CS). Methods: Data were collected from 379 preschool-aged participants, including 81 children with language disorders associated with clinical conditions. The 155-item MCELP-CS consists of five subscales: vocabulary comprehension (VC), sentence comprehension (SC), vocabulary naming (VN), sentence structure imitation (SSI), and story narration (SN). The scale was used to measure the receptive and expressive language abilities of children aged 3–5 years and 11 months. The Peabody Picture Vocabulary Test-revised (PPVT-R) was used to measure the receptive vocabulary abilities among the children (n = 338). The internal consistency, test-retest reliability, structural validity, convergent validity, and diagnostic accuracy were used to evaluate the scale. Differences between age groups were also examined using analysis of variance. Results: The MCELP-CS had high internal consistency and good test-retest reliability. Fitting indices of the two-factor model from confirmatory factor analysis (CFA), including χ², CFI, TLI, RSEAM, and SRMR, suggested that the model is consistent with the theoretical structure. Significant correlations between the MCELP-CS and PPVT-R showed a high convergent validity. In addition, the scale indicated good diagnostic accuracy in differentiating the language disorders of children with autism, cerebral palsy (CP), and hearing impairment (HI). Conclusions: The MCELP-CS is a reliable and valid diagnostic tool for language disorders of Mandarin-speaking preschool children with autism, CP, and HI. It is appropriate to collect normative data for the MCELP-CS with a large sample population of preschool children.