认知诊断计算机化自适应测量技术在心理障碍诊断与评估中的应用

The Application of Cognitive Diagnostic Computerized Adaptive Testing on Diagnosis and Assessment of Psychological Disorder

以现代测量理论为基础,该文尝试将认知诊断与计算机化自适应测验2项新技术应用于心理障碍(抑郁症)的诊断与测评,一方面探讨新技术在抑郁症诊断中的科学性与合理性,另一方面开发基于认知诊断计算机化自适应测验技术的抑郁症测评工具(简记为CD-CAT-D).研究共调查被试2 492人,经大样本数据标定及测量学指标筛选,题库最终保留136题;研究结果表明:在认知诊断理论框架下,CD-CAT-D具有较高的诊断分类一致性信度;若以PHQ-9量表作为效标,CD-CAT-D具有较理想的收敛效度和效标关联效度;同时,CD-CAT-D的灵敏度与特异度平均在0.850左右,以及AUC指标在0.80~0.90之间,这些都表明CD-CAT-D具有较理想的预测效果.这为心理障碍的诊断与评估提供了一种全新的方法和技术支持.

Based on modern measurement theory,it is attempted in this paper to apply two new technologies,cognitive diagnosis and computerized adaptive testing to the diagnosis and evaluation of mental disorders (depression).On the one hand,the science and rationality of new technologies are explored in the diagnosis of depression.On the other hand,a depression assessment tool based on cognitive diagnostic computerized adaptive testing (called as CD-CAT-D) is constructed.After model estimation of the data from 2 492 subjects,the items are screened by several Psychometric indicators,and 136 questions are finally retained in the question bank.The research results also show that under the framework of cognitive diagnosis theory,CD-CAT-D has high diagnostic classification consistency reliability.If the PHQ-9 scale is used as the criterion,CD-CAT-D has ideal convergent validity and criterion-associated validity.Meanwhile,the sensitivity and specificity of CD-CAT-D are around 0.850 on average and the AUC index is between 0.80 and 0.90 indicating that CD-CAT-D has good predictive validity.In short,a new method and technical support are provided for the diagnosis and evaluation of mental disorders.