Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Harman One Factor test

From   Marco Ercolani <>
To   "" <>
Subject   st: Harman One Factor test
Date   Thu, 30 Aug 2012 06:46:10 +0100

Dear Statalisters,

I wish to learn how to apply a simple "Harman One Factor test" to
identify "common methods variance" (and do so in Stata not SPSS).

I know that the use of this test has been discouraged in a previous thread:
but let us assume I wish to learn it for didactic reasons.

For argument's sake, let the measured (log) wage be made up of 3 parts:
lnWage = Trait_Variance + Common_Method Variance + Error_Variance
and we wish to check if Common Method Variance is a significant feature
in a wage regression.

To make things more concrete, let us assume that I set up the following
Stata dataset and analysis:
-begin program---------------------------------------------------------
webuse nlsw88.dta, clear
generate lnWage=ln(wage)
regress lnWage age married collgrad south c_city union hours tenure
* Do I then run the “factor” command? How do I interpret the results?
factor lnWage age married collgrad south c_city union hours tenure, pcf
-end program-----------------------------------------------------------

Please tolerate my ignorance on this matter, hence the post.
Simple explanations most welcome.


Dr Marco G. Ercolani
Department of Economics
University of Birmingham
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index