Development of a Mandarin expressive and receptive vocabulary test for children using cochlear implants

Abstract

Cochlear implants provide children who have profound hearing loss access to sounds and improve their speech perception, speech production and language skills. Outcomes on language development are useful for counseling and in planning habilitation programs. However, such research in mainland China is scarce due to a lack of standardized language measurement materials. This study aims at developing a vocabulary measurement named the Mandarin Expressive and Receptive Vocabulary Test (MERVT) for Mandarin-speaking children with cochlear implants. In Phase 1, items were selected from the existing vocabulary tests, corpus, story books and TV programs that were suitable for children with normal hearing aged 1 year and 6 months to 3 years and 11 months. These items were rated for age-appropriateness by kindergarten teachers. Based on the teachers’ ratings, 200 words were selected to form the initial vocabulary test. In Phase 2, responses from 102 normal-hearing preschool children were used for item analysis and identifying items with appropriate difficulty and discriminatory ability. Finally, 161 items with difficulty from 0.15 to 0.85 and discrimination greater than 0.25 were kept and ranked in an order of ascending difficulty to form the final test. In Phase 3, 245 normal-hearing children aged 1 year and 6 months to 3 years and 11 months were recruited to establish norms and evaluate psychometric properties of the test. In the last phase, 29 children with cochlear implants were recruited in the study to evaluate the validity of the MERVT for measuring the vocabulary development of children with cochlear implants. The results of the 245 normal-hearing children show that there is good internal consistency among items in both the expressive and receptive subtests (α = .83-.97). Item selection from language materials for Mandarin-speaking preschool children, the ratings by kindergarten teachers and item analysis based on the responses of target children provided evidence of content validity. Significant relationships between the MERVT and the Gesell Developmental Scale (rs = .37-.60, all ps<.05), significant correlations between subtest scores and the total score (rs = .95-.97, all ps<.01), and the gradual increase in mean subtest scores with age provided evidence of construct validity. When the MERVT was administered on a group of children with cochlear implants, significant correlation between scores on the MERVT and the parents’ report—the Chinese version of the MacArthur Communicative Development Inventory (CCDI)—(rs = .58-.70, all ps＜.05), provided convergent evidence of the construct validity of the MERVT. The growth of the MERVT scores with the increased duration of cochlear implant use and chronological age offered additional evidence of the construct validity. In conclusion, with good reliability and strong validity, the MERVT could be used for evaluating vocabulary development in young children. The test has also been proven to be a valid tool for measuring and monitoring the vocabulary development of children with cochlear implants. In addition, the MERVT and the CCDI can be used together in a hierarchical set of battery tests for measuring the vocabulary development of children with cochlear implants based on the age equivalents.