華語嬰幼兒溝通發展量表之編製與應用

The Standardization and Application of Mandarin-Chinese Communicative Developmental Inventory for Infants and Toddlers

研究目的：發展一套符合心理計量標準，且適用於國內語言環境的「華語嬰幼兒溝通發展量表」，以評量嬰幼兒語言與溝通發展，做為早期篩選語言發展遲緩的參考。研究方法：參照「MacArthur-Bates Communicative Development Inventory」的架構，結合華語溝通發展的文獻，以家長報告與結構化量表的方式蒐集8-36個月華語嬰幼兒溝通發展資料，並編製成量表。嬰兒版評量8-16個月嬰兒的溝通手勢動作和詞彙理解／表達能力，幼兒版評量16-36個月幼兒的詞彙和語法表達能力。研究結果：經初試和預試，分層隨機選取台灣地區共2654名嬰幼兒，以月齡為單位，建立各年齡組溝通發展的百分等級常模。信度考驗顯示，本量表具有頗佳的隔週再測信度、觀察者間信度及內部一致性。在效度部分，各項發展分數隨年齡增長而增加，且與「嬰幼兒綜合發展測驗」之語言分量表和實驗室觀察之間也有不錯的同時效標關聯效度，其追蹤六個月的預測效度也不錯。研究結論：整體而言，本量表具有篩選早期語言發展遲緩幼兒的臨床價值，也對檢視嬰幼兒語言發展理論具有重要貢獻。

Purpose: The primary goal of this study was to develop an age-appropriate tool (Mandarin-Chinese Communicative Development Inventory, MCDI) for assessing the language and communication development of Mandarin-speaking infants and toddlers aged from 8 to 36 months in Taiwan. Methods: The adaptation of the MCDI (Taiwan version) was based on the structure of the MacArthur-Bates Communicative Development Inventory for English-speaking children. The technique of parental report was applied to describe the developmental course and individual variability in core features of language and communication development. This inventory consisted of 2 separate forms. The MCDI/Words and Gestures (Infant Form), which is designed for 8- to 16-month-old infants, assesses vocabulary comprehension/production, phrase comprehension and the use of communication gestures. The MCDI/Words and Sentences (Toddler Form), which is designed for 16- to 36-month-old toddlers, assesses vocabulary production and syntactic complexity. Results: Various precautions were taken during the course of item collection, item analysis, pilot testing, standardization, reliability and validity testing to assure the psychometric properties of this inventory. The month-to-month norm data were collected from a stratified random sample of 2,654 8-36 month-old children in Taiwan. The results of the reliability and validity analyses demonstrated this inventory meets the psychometric criteria, which include a high one-week test-retest reliability, inter-rater reliability, internal consistency, criterion related validity (with the language subscale of the Comprehensive Developmental Inventory for Infants and Toddlers, CDIIT, and laboratory observations), and 6-month predictive validity. Conclusions: This cost-effective inventory with adequate psychometric properties can serve as a clinically valuable tool to screen and identify infants and toddlers with language developmental delay. In addition, results of this study contribute to assessment of theoretical issues in early language development.