摘要
目的使用项目反应理论(item response theory,IRT)对SF-36量表进行分析,探讨更适合于量表分析的方法。方法采用横断面研究。使用SF-36量表调查526例确诊的鼻咽癌患者,使用部分评分模型对反映生理健康的21个条目进行分析:计算PSI(person separation index),估计阈值参数和生存质量,对性别、年龄、婚姻等进行项目功能差异(DIF)分析。结果 PSI=0.85,整个模型的拟合效果好;除了条目3i、3j、8的阈值参数顺序出现颠倒外,其余条目均符合理论假设;DIF分析结果显示,所有的条目都没有非一致性DIF,条目4a在性别上存在一致性DIF,条目3g、10a在病人来源上存在一致性DIF。结论 SF-36生理健康的21个条目基本满足单维性、局部独立性的要求,可以放在一起进行IRT分析。IRT能够克服等级资料的限制,能力参数与阈值参数定义在同一个标尺上,而且存在参数不变性的优点,更加适合量表的分析评价。
Objective To explore the effectiveness of item response theory（IRT） on the evaluation of QOL by SF-36 scale.Methods The cross-sectional study was used.526 NPC（nasopharyngeal carcinoma） patients were face-to-face interviewed with the SF-36 by the well-trained investigators in Cancer Centre of Sun Yat-sen University.Partial credit model（PCM） was used to analyze 21 items indicating physical health from SF-36.The content included： to calculate the person separation index（PSI）,to get the item threshold parameter and QOL ability,and know whether these 21 items had DIF among different gender,age,marriage and so on.Results PSI was equal to 0.85,which indicated the fitting effect of the whole model was good.Except for the item 3i,3j,and item 8,the threshold order of all other were consistent with the theoretical hypothesis.All the items didn＇t have inconsistent DIF.The item 4a was diagnosed as uniform DIF between different gender,and the item 3g and 10a were diagnosed as uniform DIF between different sources.Conclusion These 21 items almost met single dimension and local independent,and they were suitable for IRT analysis.Compared with classical test theory,IRT could overcome the limitation of ordered data,and has the invariance of parameter,and with the ability parameter and the threshold parameter being in the same axis.IRT is a more suitable method to analyze the scale.
出处
《中国慢性病预防与控制》
CAS
北大核心
2010年第5期444-447,共4页
Chinese Journal of Prevention and Control of Chronic Diseases
基金
广东省科技计划项目(2006B36002026)
关键词
项目反应理论
鼻咽癌
生存质量
SF-36
Item response theory
Nasopharyngeal carcinoma
Quality of life
SF-36
