Department of Mathematics, Applied Statistics program, West Chester University, West Chester, Pennsylvania 19383, USA.
Psychother Res. 2009 Jul;19(4-5):438-52. doi: 10.1080/10503300902849475.
The authors previously reviewed the basic elements and steps to building multilevel models (MLMs) for longitudinal data typically found in psychotherapy research. The objective of this article is to focus on complexities associated with the MLM for longitudinal data analysis in psychotherapy research, which may result in proper use or misuse of the modeling structure. To do so, the authors illustrate complex scenarios and discuss issues in the implementation and interpretation of the MLM: (a) impact of missing data in the MLM, (b) determination of the complexity of the covariance structure and its implication on model interpretation, (c) issues with centering, (d) model diagnostics for MLM, (e) model formation, including implementation dependent on the treatment of time and distribution of outcome, and (f) model estimation. The authors also present data from psychotherapy research settings as examples of these complex situations. Finally, they offer some caveats and advice for recognizing these complexities and proper procession to ensure accurate implementation of the MLM and interpretation of the results.
作者之前回顾了构建常用于心理治疗研究中纵向数据的多层次模型 (MLM) 的基本要素和步骤。本文的目的是关注与心理治疗研究中纵向数据分析相关的 MLM 复杂性，这可能导致对模型结构的正确或错误使用。为此，作者举例说明了复杂的情况，并讨论了 MLM 中实施和解释的问题：(a) MLM 中缺失数据的影响，(b)协方差结构的复杂性及其对模型解释的影响的确定，(c)中心化问题，(d) MLM 的模型诊断，(e)模型形成，包括取决于治疗时间和结果分布的实施，以及 (f)模型估计。作者还提供了来自心理治疗研究环境的数据作为这些复杂情况的示例。最后，他们提出了一些注意事项和建议，以识别这些复杂性并进行正确处理，以确保 MLM 的准确实施和结果的正确解释。
