Beetham Kassia S, Bundy Anita C, Wyver Shirley, Ragen Jo, Villeneuve Michelle, Tranter Paul, Naughton Geraldine
Kassia S. Beetham, PhD, is Lecturer, School of Behavioural and Health Sciences, Australian Catholic University, Brisbane, Queensland, Australia. At the time of the study, she was Research Associate, Faculty of Health Sciences, The University of Sydney, Sydney, New South Wales, Australia.
Anita C. Bundy, ScD, is Professor and Department Head, Occupational Therapy, Colorado State University, Fort Collins, and Honorary Professor, Faculty of Health Sciences, The University of Sydney, Sydney, New South Wales, Australia;
Am J Occup Ther. 2019 Jul/Aug;73(4):7304205100p1-7304205100p10. doi: 10.5014/ajot.2019.029561.
A reliable, valid, and readily usable method of measuring coping is essential for occupational therapy practitioners and researchers working with children with developmental disabilities. The aim of this study was to examine evidence for the construct validity and internal and test-retest reliability of the Coping Inventory (CI), a 48-item survey designed for use with children.
School staff (N = 39) completed CIs for 79 students with developmental disabilities age 5-13 yr (mean = 8.0, standard deviation = 2.0) who were primarily boys (n = 52; 68%) with autism spectrum disorder (n = 40 of 47 diagnoses provided). We used Rasch analysis to examine construct validity, internal reliability, and possible redundancy of items and intraclass correlations, Pearson correlations, Bland-Altman plots, and t tests were used to examine test-retest reliability.
Rasch analysis provided evidence of construct validity: All items correlated positively with the overall measure; data from 94% of items conformed with Rasch expectations, and the items form a logical hierarchy. Intraclass correlations revealed an average coefficient of .96, providing evidence of test-retest reliability.
The CI demonstrated good evidence for construct validity and internal and test-retest reliability. Redundancy of items suggests that some items could be eliminated after further research.
对于从事发育障碍儿童工作的职业治疗师和研究人员而言，一种可靠、有效且易于使用的应对方式测量方法至关重要。本研究的目的是检验应对量表（CI）的结构效度、内部效度以及重测信度，该量表是一项为儿童设计的包含48个条目的调查问卷。
学校工作人员（N = 39）为79名年龄在5至13岁（平均 = 8.0，标准差 = 2.0）的发育障碍学生完成了应对量表调查，这些学生主要为男孩（n = 52；68%），其中47例已确诊病例中有40例患有自闭症谱系障碍。我们使用拉施分析来检验结构效度、内部信度以及项目可能存在的冗余，使用组内相关系数、皮尔逊相关系数、布兰德 - 奥特曼图和t检验来检验重测信度。
拉施分析提供了结构效度的证据：所有项目与总体测量结果呈正相关；94%的项目数据符合拉施预期，且这些项目形成了一个逻辑层次结构。组内相关系数显示平均系数为0.96，为重测信度提供了证据。
应对量表在结构效度、内部效度和重测信度方面都有充分的证据。项目冗余表明在进一步研究后一些项目可以被剔除。
