Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: RE: factor score option

From   Nick Cox <>
To   "''" <>
Subject   st: RE: factor score option
Date   Thu, 10 Feb 2011 16:17:26 +0000

This came up before and there were no replies. Why? I guess because we don't know what Jet's project or discipline points to, default scores, Bartlett scores or regression scores. 

Either way, if you don't have an idea yourself, why not just use -pca-? Alternatively, in practice if factors are well determined the scores calculated by very different methods should give very similar results. Calculate them all and check for yourself. 


   I have a question regarding the "predict" after factor analysis. I
first just use "predict X1 X2", it gives me one set of factor scores;
then, I tried to use "predict X3 X4, r" it gives me another set of
factor scores; I tried a third "predict X5 X6, b", I got a third set.
All of them are different in mean, min and max. What is the default
setting for "predict"? Which one would you recommend? thank you.

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index