Latent Transition Analysis With Random Intercepts (RI-LTA)
Latent Transition Analysis With Random Intercepts (RI-LTA)
复制标题
DOI：
10.1037/met0000370
复制
发表时间：
2022-02-01
影响因子：
7
通讯作者：
Asparouhov, Tihomir
中科院分区：
文献类型：
--
作者：
Muthen, Bengt;Asparouhov, Tihomir
随着时间的推移，潜在类别的建模是心理学中研究幸福或抑郁等心理状态随时间发展的常用方法。潜在转移分析是用于此目的的一种众所周知的方法。这里提出了一种更好的统计方法，它可以更好、更正确地表示数据，评估随时间的变化和稳定性。这种新方法改变了对心理变化过程的解释。早期LTA的发现需要重新审视。本文证明了常规的LTA模型是不必要的限制，并且可以随时使用替代模型，该模型通常更适合数据，可以更好地估计转移概率，并从数据中提取新信息。通过允许模型中的随机截距变化，可以将受试者之间的变化与受试者内部潜在类别随时间的变化分离开来，从而更清晰地解释数据。通过对文献中两个实例的分析，证明了随机截距LTA的优点。模型变化包括移动-停留分析、测量不变性分析和协变量分析。
Translational Abstract Modeling with latent classes over time is a common approach in psychology when studying the development of for example mental states of happiness or depression over time. Latent transition analysis is a well-known approach for this purpose. A better statistical approach is presented here which represents the data better and more correctly assesses change and stability over time. Interpretations of psychological change processes are changed by this new methodology. Earlier LTA findings need to be revisited.This article demonstrates that the regular LTA model is unnecessarily restrictive and that an alternative model is readily available that typically fits the data much better, leads to better estimates of the transition probabilities, and extracts new information from the data. By allowing random intercept variation in the model, between-subject variation is separated from the within-subject latent class transitions over time allowing a clearer interpretation of the data. Analysis of two examples from the literature demonstrates the advantages of random intercept LTA. Model variations include Mover-Stayer analysis, measurement invariance analysis, and analysis with covariates.