School of Psychology, University of Tasmania, Australia.
Psychol Assess. 2012 Mar;24(1):1-10. doi: 10.1037/a0024966. Epub 2011 Aug 22.
In the study, the authors examined the measurement (configural, factor loadings, thresholds, and error variances) and structural (factor variances, covariances, and mean scores) invariance of the Children's Depression Inventory (CDI; Kovacs, 1992) across ratings provided by clinic-referred children and adolescents with (N = 383) and without (N = 412) depressive disorders. Multiple-group confirmatory factor analysis of the Craighead, Smucker, Craighead, and Ilardi (1998) CDI model supported full measurement invariance and invariance for structural variances and covariances. Invariance for thresholds was also supported by multiple indicators multiple causes (MIMIC) procedures that controlled for the effects of age; sex; and the presence or absence of anxiety disorders, attention-deficit/hyperactivity disorder, and oppositional defiant/conduct disorders. The MIMIC analyses showed that for latent mean scores, the group with depressive disorders had higher scores, with at least medium effect sizes, for Self-Deprecation and Biological Dysregulation. The theoretical, psychometric, and clinical implications of the findings are discussed.
在这项研究中，作者检查了儿童抑郁量表（CDI；Kovacs，1992）的测量（组态、因子负荷、阈限和误差方差）和结构（因子方差、协方差和平均分数）不变性，这些不变性是通过临床转介的有（N=383）和没有（N=412）抑郁障碍的儿童和青少年提供的评分得出的。Craighead、Smucker、Craighead 和 Ilardi（1998）CDI 模型的多组验证性因子分析支持完全测量不变性和结构方差和协方差的不变性。多指标多原因（MIMIC）程序控制年龄、性别以及焦虑障碍、注意缺陷/多动障碍和对立违抗性/品行障碍的存在与否，也支持阈限不变性。MIMIC 分析表明，对于潜在的平均分数，有抑郁障碍的组在自我贬低和生物失调方面的得分更高，具有至少中等的效应大小。讨论了研究结果的理论、心理测量和临床意义。
