教育生物学杂志2024,Vol.12Issue(6) :421-426.DOI:10.3969/j.issn.2095-4301.2024.06.001

儿童孤独症评定量表家长版的信度与效度分析

Reliability and validity analysis of parental version of childhood autism rating scale

李亦诚 霍言言 张媛媛 仇晓艳 马晨欢 殷安欣 刘巧云 王健 罗洁鸣 李天舒 徐敬 周洁 王燕 陶炳铜 孟凡萍 王瑜 蒋泓 陈津津
教育生物学杂志2024,Vol.12Issue(6) :421-426.DOI:10.3969/j.issn.2095-4301.2024.06.001

儿童孤独症评定量表家长版的信度与效度分析

Reliability and validity analysis of parental version of childhood autism rating scale

李亦诚 1霍言言 2张媛媛 2仇晓艳 2马晨欢 2殷安欣 3刘巧云 4王健 5罗洁鸣 6李天舒 6徐敬 7周洁 8王燕 9陶炳铜 10孟凡萍 11王瑜 2蒋泓 3陈津津2
扫码查看

作者信息

  • 1. 上海交通大学医学院附属儿童医院上海市儿童医院儿童保健科(中国 上海 200062);上海交通大学医学院附属第一人民医院(中国 上海 200080)
  • 2. 上海交通大学医学院附属儿童医院上海市儿童医院儿童保健科(中国 上海 200062)
  • 3. 复旦大学公共卫生学院(中国 上海 200436)
  • 4. 华东师范大学康复科学系(中国 上海 200062)
  • 5. 上海市静安区妇幼保健所(中国 上海 200072)
  • 6. 上海市黄浦区妇幼保健所(中国 上海 200010)
  • 7. 上海市杨浦区妇幼保健所(中国 上海 200082)
  • 8. 上海市普陀区妇婴保健院(中国 上海 200062)
  • 9. 同济大学附属东方医院胶州医院(中国胶州 266318)
  • 10. 南阳市第二人民医院(中国 南阳 473009)
  • 11. 成都西南儿童医院(中国 成都 610047)
  • 折叠

摘要

目的 检验儿童孤独症评定量表(childhood autism rating scale,CARS)家长版的可靠性和有效性.方法 在上海市 3 个区招募并随机抽取 319 名儿童及其家长参与研究.基于小组讨论和德尔菲专家函询方法对原版CARS进行注释,使其易于理解并适合家长填写,形成CARS家长版.在参加 2 个课时的线上课堂后,家长根据孩子的日常表现填写量表,完成线上评估.对量表的区分度、结构效度、效标关联效度、同质性信度、重测信度和评价者间信度进行统计检验.结果 CARS家长版 15 个条目的项目区分度为 0.588~0.801.探索性因子分析示其KMO值为 0.788,Bartlett's球形检验显示P<0.001,提取 4 个公因子后累积方差解释率为 61.74%,各条目的因子载荷系数为 0.443~0.801;验证性因子分析支持 4 因子模型(χ2/df=2.834,RMSEA=0.107,GFI=0.842,CFI=0.814,AGFI=0.774,IFI=0.818,TLI=0.767).该量表与改良婴幼儿孤独症量表、儿童孤独症行为量表和孤独症诊断观察量表的效标关联效度分别为 0.41、0.69 和 0.43(P<0.01).量表的Cronbach's α系数、分半信度、重测信度和评分者间信度分别为 0.91、0.90、0.81 和 0.71.结论 CARS家长版满足测量学标准,具有良好的信效度,可用于孤独症谱系障碍儿童的居家早期筛查.

Abstract

Objective To examine the reliability and validity of the parent version of the childhood autism rating scale(CARS).Methods In 3 districts of Shanghai,319 children and their parents were recruited and randomly selected to participate in the study.Based on group discussion and Delphi expert inquiry method,the original version of CARS was annotated to make it easy to understand and suitable to fill out for parents,forming the parent version of CARS.After attending two online classes,parents filled out the scale according to their children's daily performance and completed an online assessment.The discriminant validity,structural validity,criterion-related validity,homogeneity reliability,test-retest reliability,and inter-rater reliability of the scale were statistically tested.Results The item range of discriminant validity for the 15 items of the parent version of CARS was 0.588-0.801.Exploratory factor analysis showed a KMO value of 0.788 and a Bartlett's sphere test with P<0.001.After extracting 4 common factors,the cumulative variance contribution rate was 61.74%,and the factor loading of each item was 0.443-0.801.Confirmatory factor analysis supported the 4-factor model(χ2/df=2.834,RMSEA=0.107,GFI=0.842,CFI=0.814,AGFI=0.774,IFI=0.818,TLI=0.767).The criterion-related validity of the scale was 0.41,0.69,and 0.43(P<0.01)when compared to the modified checklist for autism in toddlers(M-CHAT),autism behavior checklist(ABC),and autism diagnostic observation schedule(ADOS).The Cronbach's α coefficient,split-half reliability,test-retest reliability,and inter-rater reliability of the scale were 0.91,0.90,0.81,and 0.71,respectively.Conclusion The parent version of CARS meets the measurement standards and demonstrates good reliability and validity,making it suitable for early home screening of children with autism spectrum disorder.

关键词

孤独症谱系障碍/早期筛查/儿童孤独症评定量表/家长版/信度/效度

Key words

autism spectrum disorder/early screening/childhood autism rating scale/parental version/reliability/validity

引用本文复制引用

出版年

2024
教育生物学杂志
上海交通大学医学院附属新华医院

教育生物学杂志

CSTPCD
影响因子:0.225
ISSN:2095-4301
段落导航相关论文