透過您的圖書館登入
IP:13.58.137.218
  • 期刊

多層次模式的交互作用與脈絡變數的飽和模式分析:以組織氣氛知覺對工作滿意的影響為例

Saturated Modeling of the Interaction with Contextual Variable in Multilevel Modeling: An Example of Effect of Perceived Organizational Climate on Job Satisfaction

摘要


多層次模式是組織與人管研究重要的研究策略,其主要的價值之一是可針對不同類型的交互作用(同階層與跨階層)進行檢驗,進而提供多元化的調節效果分析結論。本文除了檢視多層次模式中各種交互作用的意義與特性,並針對脈絡變數的效果與脈絡交互作用,提出飽和混合模式策略來進行估計,同時也針對組平減與總平減等不同的中心化策略的影響進行檢驗。實徵資料來自於24家企業664名員工的組織創新氣氛與滿意度測量數據,並以個體層次的性別與總體層次的產業別為調節變數,進行飽和混合模式分析。實徵數據的分析發現,個體層次解釋變數經過組平減並置入總體層次的截距與斜率方程式,對於各項參數估計與模式適配檢驗具有最理想的結果。低階交互作用會被高階解釋變數調節,使得調節效果的分析相當複雜,本文除了針對不同形式的交互作用平減操作提出建議之外,也對多層次迴歸的相關議題進行討論。

並列摘要


The major advantage of multilevel linear modeling (MLM) which is widely used in the organizational and human resource management study is the examination of different forms of interaction, i.e. within-level as well as cross-level interaction. As a result, moderation effects present in the different forms of interaction can be examined. The present study reviews the properties of the different interaction in MLM, along with a particularly interest in the treatment of contextual variable as well as self-served interaction. Two strategies of centering of predictor, entitled as group-mean centering and grand-mean centering are discussed and illustrated by the partial data selected from author's previous study consisted of 664 employees sampled from 24 different organizations. The moderation effects of gender and industrial type on the job satisfaction predicted by the perceived organizational climate were examined by using saturated mixed modeling. Results revealed that the group-mean centering with replacement into the intercept and slope equation at macro-level was the best estimates of fixed effects as well as random effects. Implications and limitations of the multilevel moderated multiple regression were discussed in the final section.

參考文獻


林鉦棽、彭台光(2006)。多層次管理研究:分析層次的概念、理論和方法。管理學報。23(6),649-675。
邱皓政、郭建志、劉兆明()。,未出版。
溫福星、邱皓政(2009)。多層次模型方法論:階層線性模式的關鍵議題與試解。臺大管理論叢。19(2),263-294。
Aguinis, H.(2004).Regression analysis for categorical moderators.New York:Guilford Press.
Aguinis, H.,Culpepper, S. A.(2015).An expanded decision making procedure for examining cross-level interaction effects with multilevel modeling.Organizational Research Methods.18,155-176.

被引用紀錄


劉胤含、胡亞平、邱紹群(2023)。疫情期間混合任務環境對員工績效影響之研究管理資訊計算12(2),103-118。https://doi.org/10.6285/MIC.202309_12(2).0008

延伸閱讀