| Summary: | 碩士 === 國立成功大學 === 行為醫學研究所 === 95 === 【Purpose】
The concurrent criterion-related validity of the Chinese Child Development Inventory (CCDI) was assessed with a mental development index (MDI) and psychomotor development index (PDI) measured by the Bayley scales of Infant Development-II (BSID-II). In addition, the cut-off points of standard scores in eight subscales were investigated for identifying developmental delays. And the diagnostic accuracy indicators of standard scores and developmental ages were both further examined and compared.
【Research Questions】
1. Investigated the correlations between CCDI and BSID-II.
2. With MDI as criterion, surveyed the diagnostic accuracy indicators of each cut-off point in standard scores for eight subscales, and graphed out the Receiver Operating Characteristic (ROC) curves to determine the most efficient cut-off point.
3. With PDI as criterion, surveyed the diagnostic accuracy indicators of each cut-off point in standard scores for General Development, Gross Motor and Fine Motor; and graphed out the Receiver Operating Characteristic (ROC) curves to determine the most efficient cut-off point.
4. With MDI and PDI on BSID-II as criterion, investigated and compared the diagnostic accuracy indicators between 20% and 30% below-age-level cutoff points.
【Method】
The present study employed a cross-sectional study design. 93 subjects were purposive drawn from development clinic, premature follow-up clinic and volunteers in community by advertisement. 40 girls and 53 boys ranged from 12 to 42 months, with a mean chronological age of 26.69 months. When BSID-II was administered to children, the parents were asked to answer SES questionnaire and CCDI.
【Results】
1. Pearson correlation procedures revealed significantly moderate to high correlations between MDI and the General Development (GD) and Conceptual Comprehension (CC) of CCDI, and a significantly moderate correlation between PDI and the Gross Motor (GM).
2. In crosstabulation analyses, with MDI as criterion, the most efficient cut-off point is GD standard score <= 85, which with good sensitivity (80.8%) and excellent specificity (90.5%).
3. In crosstabulation analyses, with PDI as criterion, the overall accuracy of GM, FM and GD was significantly higher than random guess. Considered the diagnostic accuracy indicators among each cutoffs, the relatively efficient cut-off point is GM standard score <= 8, with sensitivity=81% and specificity =77.8%.
4. The screening efficiency of 20% below-age-level cutoff point with MDI was optimized, and the hit rates of GD, CC and Expression Language (EL) were all above 80%.
【Discussion】
A cutoff GD standard score of 85 maximized accurate classification of delayed and normal development status, with sensitivity and specificity rate both above 80%. The overall accuracy of GM and FM were optimized to predict PDI, but in each cut-off point, the sensitivity and specificity were not above 80% at the same time. Therefore, CCDI should combine other development instruments for identifying between normal or delay of motor development. In addition, the screening efficiency of 20% below-age-level cutoff point with MDI was optimized than 30% below-age-level cutoff point. This study shows that the new norm has good criterion-related validity, for both practical and research-based screening tool for the development of preschool children.
|
|---|
