部分複相關係數在管理研究之探究－控制變數與軟體配合應用

On Partial Multiple Correlation Coefficients in Managerial Studies -- The Combined Application of Control Variable and Software

龔千芬(Chien-Feng Kung)

《人力資源管理學報》 17卷2期 (2017/06) Pp. 73-92

https://doi.org/10.6147/JHRM.2017.1702.04

摘要

控制變數在管理、組織、與心理各個領域皆被研究者廣泛應用，Organizational Research Methods在2012年更以專刊介紹控制變數相關議題，可見控制變數在研究中應用普遍與重要性。控制變數應用於研究者在控制基本變數下，專注於主要變數對於反應變數的結果，就統計方法而言，部份複相關分析較實用，但卻僅少數學者進行相關議題的研究，因為部份複相關分析理論結構十分複雜，形成重要關卡，更缺乏完整統計軟體支援。本研究著重利用部份複相關分析於假設檢定、檢定力計算、以及樣本數等議題，利用Excel介面的親和性與普及性，提供全面性且實務性的介紹作為研究規劃與分析之用。最後，配合個案詳細的說明如何運用此軟體於規劃研究決策或教學展示上。

關鍵字

部分複相關係數 ； 控制變數 ； 檢定力計算 ； 假設檢定 ； 樣本數決定

並列摘要

Control variables are widely used by researchers in various fields, including management, organization and psychology. In 2012, Organizational Research Methods made an exclusive description of the topics about control variables, which demonstrated the ubiquity and importance of control variables in researches. Control variables are applied when researchers control basic variables and focus on the effects of main variables on response variables. As far as the statistical methodology is concerned, the partial multiple correlation analysis is more practical than the multiple correlation analysis. Currently, only a small number of scholars are exploring the topics about the practical partial multiple correlation analysis. The reason for this is that the theoretical structure of the partial multiple correlation is very complicated, which sets a high threshold, and that there is lack of complete statistical software application to support. This research inferences exact probability density function of the partial multiple correlation coefficients, and a extensive accessibity of Microsoft Excel software, the associated computer routines for hypothesis testing, power calculation, and sample determination are developed. With the compatibility and popularity of the Excel interface, this study gives an introduction to its comprehensiveness and practicability to offer reference for research planning and analysis. Finally, a research case is cited for a detailed description of how to apply the software in planning, research and decision-making or instruction presentation.

並列關鍵字

partial multiple correlation coefficient ； control variables ； power calculation ； hypothesis testing ； sample size determination

參考文獻

Bathurst, J. E. 1927. A partial correlation schema. Journal of Applied Psychology, 11: 154-164.

Burks, B. S. 1926. On the inadequacy of the partial and multiple correlation technique. Journal of Educational Psychology, 17: 625-630.

Wren, F. L. 1938. The calculation of partial and multiple coefficients of regression and correlation. Journal of Educational Psychology, 29: 695-700.

Algina, J.,Keselman, H. J.,Penfield, R. J.(2008).Note on a confidence interval for the squared semipartial correlation coefficient.Educational and Psychological Measurement.68,734-741.

Algina, J.,Moulder, B. C.(2001).Sample sizes for confidence intervals on the increase in the squared multiple correlation coefficient.Educational and Psychological Measurement.61,633-649.