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]

Re: st: Generate one score for 2+ factors?

From   Maarten Buis <>
Subject   Re: st: Generate one score for 2+ factors?
Date   Thu, 27 Jun 2013 09:56:30 +0200

On Wed, Jun 26, 2013 at 7:52 PM, Nick Cox wrote:
> If you want a single construct to use you
> need really good arguments not to use Factor 1.

One situation where you might choose Factor 2 over Factor 1 can occur
when you have a set of question where some are phrased positive and
others are phrased negative. There is a tendency for respondents to
agree with whatever is being said, and the first factor may pick that
up. You'll see that when the first factor loads highly on all
positively phrased statements and has low loadings for the negatively
phrased statements (or vice versa). That would be a case where I would
consider the first factor as measuring an uninteresting artefact, and
I would move on to the next factor.

-- Maarten

Maarten L. Buis
Reichpietschufer 50
10785 Berlin
*   For searches and help try:

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