控制變數這個問題許多人問過，首先回答一個問題，你知道什麼是控制變數嗎?為何研究要放控制變數呢?理由大多是別人有用，所以我也用，因此就造成了許多控制變數的誤用。
從理論上先瞭解性別，教育程度是名義及順序尺度，控制變數其實在統計上是回歸的偏相關，也是引數，回歸的基本假設是引數與因變數必需是線性相關,性別為類別變數，跑出的值是平均值差異，而非斜率，教育程度與Y不可能為線性關係，代入也不對。
而且控制變數是理論上要和Y有顯著相關的變數，但又不是研究的重點，所以分析完後，如果控制變數與Y不顯著，表示研究設控制變數是錯的。
引用文獻出處：
Use specific, well-explored theory to drivethe inclusion of
controls, which goes beyond simple statements like, “previousresearchers used
this control” or “this variable is correlated with myoutcomes.” If you
believe that a specific relationship may becontaminating your results, this may
be justification for a control, but youshould explicit state why and defend
this decision when describing yourmethods.
Don’t control for demographic variables,e.g. race, gender,
sex, age. For example, if you find a gender difference inyour outcome of
interest, controlling for that variable may hide real variancein the outcome
that could be explained by whatever real phenomenon is causingthat difference.
In my own research are, it is not uncommon to controlfor age when examining the
effects of technology on outcomes of interest (e.g.learning). But age
does not itself cause trouble with technology;instead, underlying differences
like familiarity with technology or comfortwith technology or other
characteristics may be driving thosedifferences. Simply controlling for
age not only removes “real” variancethat should remain in the equation but also
camouflages a real relationship ofinterest.
Richard, N. L. (2011). Stats and Methods Urban Legend 2:
Control Variables Improve Your Study.
Paul, E. S., & Michael, T. B. (2011). Methodological Urban
Legends: The Misuse of Statistical Control Variables. Organizational Research Methods, 14(2),287-305.
沒有留言:
張貼留言