學齡前兒童動作發展篩檢量表之開發與驗證－初步研究

Developing and Validating a Screening Scale for Assessing Preschool Children's Motor Development: A Preliminary Study

汪家琦(Chia-Chi Wang) ； 張鑑如(Chien-Ju Chang) ； 陳柏熹(Po-Hsi Chen) ； 劉文瑜(Wen-Yu Liu) ； 陳嘉玲(Chia-Ling Chen) ； 蔡志謙(Chih-Chien Tsai) ； 傅仁煇(Ren-Huei Fu) ； 周麗端(Li-Tuan Chou) ； 朱世明(Shih-Ming Chu)

《台灣復健醫學雜誌》 43卷4期 (2015/12) Pp. 239-250

https://doi.org/10.6315/2015.43(4)04

摘要

背景與目的：國內僅針對學齡前動作發展的篩檢問卷甚少，因此本研究目的為開發兼具信度與效度之「學齡前兒童動作發展篩檢量表」（Developmental Motor Screening Scale for Preschool Children，簡稱 DMSSPC）。研究方法：本研究參考臨床動作評估工具及篩檢問卷，經由召開專家會議討論，自行研發「學齡前 兒童動作發展篩檢問卷」，分為精細動作和粗大細動作兩個部分，前者分為抓握、手部操作以及視動整 合三個向度；後者分為穩定度、移位以及身體協調三個向度。招收台北市以及新北市地區共150 位學齡 前兒童，以每12 個月為一組，共分為6組，其中24位兒童接受皮巴迪動作量表第二版(Peabody Developmental Motor Scale 2nd Edition，簡稱PDMS-2)測驗，以及72位家長於一週內進行問卷再測。本研 究使用Rasch分析，依據題目難度估計值、均方誤(mean square error，簡稱MNSQ)等做為題目適配度檢 測，進行題目修訂，信度方面，建立問卷內部一致性(Cronbach's α)和組內相關係數(Intra-class correlation coefficient, ICC)估計再測信度；效度方面，建立專家評分後的專家效度(S-CVI)，並以PDMS-2作為校標，使用皮爾森相關係數(Pearson's correlation coefficient)為檢測指標，另外，使用ANOVA 以及事後分析比 較各年齡組間差異(p＜0.01)。結果：經Rasch分析精細動作為71 題，粗大動作為95題，其難度估計精細動作介於-9.77－4.52， 粗大動作則介於-7.90－3.79。初步分析信效度，DMSSPC 具有良好的信度，內部一致性Cronbach's α皆大於0.920以上；再測信度達顯著相關(ICC=0.797－0.988，p＜0.01)；效度檢測方面，DMSSPC專家效度S-CVI為0.734－1.000。而精細動作各向度與PDMS-2呈高度相關(r=0.93－0.97, p＜0.01)，粗大動作，除了反射向度外，其餘各向度與PDMS-2呈高度相關(r=0.849－0.977, p＜0.01)。組間差異，精細動作、粗大動作各向度在5歲以前各年齡組均有顯著差異(p＜0.05)，且年齡越大表現越佳。結論：根據目前分析數據，本研究初步開發的DMSSPC具有良好信效度，且在各年齡組間亦有顯著 差異，其試題難度估計偏簡單，未來可增修問卷題目之難易度。

關鍵字

學齡前兒童(preschool children) ； 動作發展篩檢問卷(motor development questionnaire) ； 信度(reliability) ； 效度(validity) ； 試題反應理論(item response theory)

並列摘要

Background: Questionnaires available for screening the motor development of preschool children in Taiwan are few. Therefore, this study developed a Developmental Motor Screening Scale for Preschool Children (DMSSPC), and it exhibits high reliability and validity. Methods: The DMSSPC was developed on the basis of an expert meeting and a literature review of assessment tools and questionnaires for screening motor development. After the meeting, subscales of the fine and gross motor scales of the DMSSPC were determined. The fine motor scale comprised grasp, manipulation, and visual motor integration subscales, whereas the gross motor scale contained stationary, locomotion, and body coordination subscales. Data on 150 preschool children (6 months to 6 years old) from Taipei City and New Taipei City were collected, and the children were divided into 6 age groups. All caregivers were interviewed by the rater and completed the DMSSPC at the initial visit. Fifty-four caregivers completed the DMSSPC again within a week. Twenty-four children were subjected to a test involving the Peabody Developmental Motor Scale, 2nd Edition (PDMS-2). The items were revised according to the difficulty estimate and unweighted fitting (mean square error) measured by Rasch analysis. The internal consistency (Cronbach's α) and test-retest reliability (intraclass correlation coefficient [ICC]) were analyzed. For validity, the expert validity and criterion validity of PDMS-2 were analyzed through S-CVI and Pearson correlation coefficients, respectively. The differences in the DMSSPC raw data among the 6 age groups were compared by analysis of variance involving post hoc analysis. Moreover, p ＜ 0.05 was considered significant. Result: After Rasch analysis, 71 and 95 items were included in the fine motor and gross motor scales, respectively. The difficulty estimates for the DMSSPC ranged from -9.77 to 4.52 for the fine motor scale and from -7.90 to 3.79 for the gross motor scale. The DMSSPC exhibited high internal consistency (Cronbach's α ＞ 0.92) and test-retest reliability (ICC = 0.797-0.988, p ＜ 0.01). The fine motor and gross motor scales of the DMSSPC were significantly correlated with all PDMS-2 domains (r = 0.85-0.98, p ＜ 0.01), except for reflex domains. Significant differences were observed in the scores for the various subscales of the gross and fine motor scales among different age groups in children aged younger than 5 years (p ＜ 0.05). The older children had higher motor scores for all subscales, except for the body coordination subscale. Conclusion: The DMSSPC developed in this study exhibited high reliability and validity. Furthermore, significant differences were observed in various motor subscale scores among different age groups. However, the difficulty estimate of all items was simpler. In the future, we will revise the DMSSPC items for adjusting the difficulty estimates.

並列關鍵字

preschool children ； motor development questionnaire ； reliability ； validity ； item response theory

參考文獻

王佳琪、鄭英耀、劉昆夏等：以Rasch分析檢驗「多向度幽默感量表」之信效度。測驗學刊 2011；58：691-713。"

陸希平、林妍如、陳家玉等：測驗給幾分的科學考量：淺談試題反應理論。醫學教育 2005；9： 278-281。"

Lin CK, Meng LF, Yu YW, et al., Factor analysis of the contextual fine motor questionnaire in children. Research in Developmental Disabilities 2014;35:512-519."

José AB, Fundamental motor skill proficiency is necessary for children's motor activity inclusion. Motriz-Revista de Educacao Fisica 2013;19: 548-551."

Piek JP, Baynam GB, Barrett NC. The relationship between fine and gross motor ability, self-perceptions and self-worth in children and adolescents. Human Movement Science 2006;25:65-75."