Refining the assessment of disrupted maternal communication: Using item response models to identify central indicators of disrupted behavior.
Refining the assessment of disrupted maternal communication: Using item response models to identify central indicators of disrupted behavior.
复制标题
DOI：
10.1017/s0954579417001778
复制
发表时间：
2019-03
影响因子：
3.3
通讯作者：
Lyons-Ruth K
中科院分区：
文献类型：
--
作者：
Haltigan JD;Madigan S;Bronfman E;Bailey HN;Borland-Kerr C;Mills-Koonce R;Lyons-Ruth K
非典型母亲行为评估和分类工具（AMBIANCE）是一种广泛使用且经过良好验证的测量方法，用于评估亲子互动中照顾者反应的紊乱形式。然而，它需要评估大约150个行为项目，从录像带和大量的培训到代码，因此在大多数临床环境中使用它是不切实际的。因此，本研究的主要目的是利用潜在特质项目反应理论（IRT）模型，确定一套精简的行为指标，这些指标是AMBIANCE编码系统的核心。先前用AMBIANCE编码的观察到的母婴互动数据来自北美和欧洲的实验室（N = 343）。使用2参数logistic IRT模型，确定了45个AMBIANCE项目的精简集。初步的收敛效度和判别效度与使用全套AMBIANCE指标分配的产妇沟通中断分类、婴儿依恋紊乱和产妇敏感性有关。研究结果支持了改进的项目集的结构效度，为开发一种简短的筛查方法来筛查产妇沟通障碍开辟了道路。本文讨论了IRT模型在临床量表的改进及其在发展精神病理学中连接临床和研究目标的潜力。
The Atypical Maternal Behavior Instrument for Assessment and Classification (AMBIANCE) is a widely used and well-validated measure for assessing disrupted forms of caregiver responsiveness within parent–child interactions. However, it requires evaluating approximately 150 behavioral items from videotape and extensive training to code, thus making its use impractical in most clinical contexts. Accordingly, the primary aim of the current study was to identify a reduced set of behavioral indicators most central to the AMBIANCE coding system using latent-trait item response theory (IRT) models. Observed mother–infant interaction data previously coded with the AMBIANCE was pooled from laboratories in both North America and Europe (N = 343). Using 2-parameter logistic IRT models, a reduced set of 45 AMBIANCE items was identified. Preliminary convergent and discriminant validity was evaluated in relation to classifications of maternal disrupted communication assigned using the full set of AMBIANCE indicators, to infant attachment disorganization, and to maternal sensitivity. The results supported the construct validity of the refined item set, opening the way for development of a brief screening measure for disrupted maternal communication. IRT models in clinical scale refinement and their potential for bridging clinical and research objectives in developmental psychopathology are discussed.