閱讀全文 購買本期 | |
篇名 |
應用線性混合效果模式於建立多層縱向資料的模式之實例研究
|
---|---|
並列篇名 | Application of Linear Mixed-Effects Models in Multilevel Longitudinal Modeling: A Case Study |
作者 | 葛湘瑋 |
中文摘要 | 縱向資料通常用於研究個人的成長、發展及個人的改變。這種形式的資料通常包含了同一個受試者在不同時間點上重複的接受測量。多變量分析和重複測量變異數分析常用來分析縱向資料。然而利用這兩項統計方法在分析縱向資料上有它的限制。縱向資料通常需要結構性的共變異數模式,殘差通常含有異質性和相依性;資料通常也屬於多層資料,重複測量是第一層,受試者是第二層。這項研究的主要目的是探討使用線性混合效果模式來建立縱向資料的模式的情形,同時也包含了如何建立模式的步驟。本研究以視覺搜尋資料當作實例分析來顯示使用線性混合效果模式在分析縱向資料的優點,同時也包括了利用線性混合效果模式在分析縱向資料的建議與討論。 |
英文摘要 | Longitudinal data are used in the research of growth, development, and change. Such data consist of the same subjects repeatedly measured over time. Multivariate analysis of variance and repeated measures ANOVA have historically been the most widely used tools for analyzing longitudinal data. However, there are some limitations in utilizing these two methods in analyzing longitudinal data. The longitudinal data typically need some structured covariance models, and the residual errors often exhibit heteroscedasticity and dependence. They also possess a hierarchical data structure in the sense that the repeated measurements are viewed as a separate level nested within an individual. The main objective of this study was to investigate hierarchical linear models/linear mixed-effects models for longitudinal data. The visual-search dataset is used to demonstrate the advantages of utilizing the methodology. Several recommenda?tions and advantages on the application of linear mixed-effects models in longitudinal modeling are discussed. |
起訖頁 | 399-419 |
關鍵詞 | 線性混合效果模式、多層線性模式、縱向資料、隨機效果、異質性、相依性、Linear mixed-effects models、Hierarchical linear models、Random effects、Longitudinal data、Heteroscedasticity、Dependence |
刊名 | 教育與心理研究 |
期數 | 200406 (27:2期) |
出版單位 | 國立政治大學教育學院 |
該期刊 上一篇
| 英文閱讀策略自我效能量表之建構 |