|
|
Second factor reverses sign of parameter |
|
Message/Author |
|
Gareth posted on Saturday, August 16, 2008 - 6:04 am
|
|
|
When regressing a two-factor CFA f1 f2 on a covariate x1, I notice that the sign of f1 ON x1 is negative. When I include f1 only, f1 ON x1 is positive. According to theory, it should be positive. Is this an example of suppression? Are there any strategies for interpreting the correct sign of the parameter? |
|
|
This does not sound correct. Please send the two outputs and your license number to support@statmodel.com. Perhaps you have different observations in the two analyses. |
|
Back to top |
|
|