| dc.contributor | 企管系 | |
| dc.creator (作者) | 胡昌亞 | |
| dc.date (日期) | 2022-10 | |
| dc.date.accessioned | 5-Jun-2026 11:56:48 (UTC+8) | - |
| dc.date.available | 5-Jun-2026 11:56:48 (UTC+8) | - |
| dc.date.issued (上傳時間) | 5-Jun-2026 11:56:48 (UTC+8) | - |
| dc.identifier.uri (URI) | https://ah.lib.nccu.edu.tw/item?item_id=182779 | - |
| dc.description.abstract (摘要) | 多層次研究是組織行為重要的研究趨勢。由於許多高層次的研究變項，是由低層次研究變項平均來代表之，故低層次變項的信度（ICC(1)）、團體大小、團體數都會影響跨層次調節效果的估計。雖然Lüdtke 等人 (2008)曾以模擬分析了解跨層次主效果的估計偏誤。但他們所使用的參數設定，組織行為的研究較難達成。此外，跨層次調節效果經常是組織行為的焦點，但該研究也沒有探討此類偏誤。有鑑於此，本研究將以模擬資料的方式，分別以多層次潛在共變分析以及多層次外顯共變分析，探討低層次變項的特質（信度、團體大小、團體數）對跨層次調節效果偏誤的影響。本研究結果將能提供研究者選擇合理的分析取向，來進行跨層次資料分析。 | |
| dc.description.abstract (摘要) | Modeling cross-level moderation effects are of great interest in organizational studies. However, the statistical properties (group size, group number, and sampling ratio) and reliability of level-1 variables (ICC (1)) can result in biases in the cross-level moderating effects of the level-2 variables. Although Lüdtke and his collogues (2008) demonstrated the potential biases of the cross-level main effect of a contextual variable, the parameters (e.g., group size and ICC) used for simulation were unrealistic for most organizational studies. Furthermore, while the cross-level moderation effect has been the focus of multilevel studies, the biases in cross-level moderation effect based on the two types of modeling approaches were not examined. To address the aforementioned issue, we will use one simulation study to examine the estimation biases of the cross-level moderation effect of reflective level-2 variables using the Multilevel Latent Covariance (MLC) approach and the Multilevel Manifest Covariance (MMC) approach. The implications and future research directions based on the findings will be discussed. | |
| dc.format.extent | 116 bytes | - |
| dc.format.mimetype | text/html | - |
| dc.relation (關聯) | 科技部, MOST109-2410-H004-091, 109.08-110.07 | |
| dc.subject (關鍵詞) | 多層次結構方程模式; 多層次潛在共變分析; 多層次外顯共變分析; 信度; 跨層次調節效果; ICC; 團隊人數; 團隊個數; 偏誤 | |
| dc.subject (關鍵詞) | Multilevel structural equation modeling; multilevel latent covariance; multilevel manifest covariance; reliability; Cross-level moderation; ICC; group size; group number; bias | |
| dc.title (題名) | 多層次跨層調節效果之估計偏誤：多層次潛在共變分析以及多層次外顯共變分析之比較 | |
| dc.title (題名) | The Biases of Cross-Level Moderation of Contextual Variables in Multilevel Modeling with the Multilevel Latent Covariance Approach and the Multilevel Manifest Covariance Approach | |
| dc.type (資料類型) | report | |
