## factor analysis question

### factor analysis question

I am using exploratory factor analysis on a rather large survey
instrument.  Several of the variables are loaded heavily (>|.3|) on
different factors. Should such variables be tossed out, or is there a
better way of dealing with the problem?
--
Adam McKee
___________________________________

work phone: 266-5460
home phone: 271-8247

### factor analysis question

I think that this feature is inevitable.
The clearest results you will become after oblique rotation.
--
OUTLIER, Consultants in Statistics

www.outlier.be

> I am using exploratory factor analysis on a rather large survey
> instrument.  Several of the variables are loaded heavily (>|.3|) on
> different factors. Should such variables be tossed out, or is there a
> better way of dealing with the problem?
> --
> Adam McKee
> ___________________________________

> work phone: 266-5460
> home phone: 271-8247

### factor analysis question

On Wed, 27 Oct 1999 10:07:02 -0500, Adam McKee

> I am using exploratory factor analysis on a rather large survey
> instrument.  Several of the variables are loaded heavily (>|.3|) on
> different factors. Should such variables be tossed out, or is there a
> better way of dealing with the problem?

- Sometimes that is a symptom of having a sample too small to
distinguish the factor structure.  If that were the case, the main
solution would be,

Get a much larger sample or use a lot fewer variables.

--

http://www.pitt.edu/~wpilib/index.html

### factor analysis question

I would not consider |.3| as a high factor loading. It's reasonable to consider
factor loadings above |.6| as high. Also, make sure that the factor loadings
you look at are the ones displayed after rotation.

### factor analysis question

Hi

> I am using exploratory factor analysis on a rather large survey
> instrument.  Several of the variables are loaded heavily (>|.3|) on
> different factors. Should such variables be tossed out, or is there a
> better way of dealing with the problem?

If you are using exploratory analysis, then there is probably not
good reason to think that each variable (whether single item or a
composite) is entirely "pure" (i.e., measures a single
dimension).  So I would take such results (presuming the loadings
are large enough, as noted by another respondent) as a hint at
the multidimensional nature of the variable.  You want to ask
yourself whether it makes sense that specific variables load on
different factors, perhaps by examining variables that are more
pure indicators of the factor.

Best wishes
Jim

============================================================================
James M. Clark                          (204) 786-9757
Department of Psychology                (204) 774-4134 Fax
University of Winnipeg                  4L05D

CANADA                                  http://www.uwinnipeg.ca/~clark
============================================================================

Hello

We have a data set which includes, for each of 96 cities, 4 estimates
of the number of drug injectors in that city.  Each of these estimates
contains error.

We ran a factor analysis to determine their commonality.

We now wish to use the results of this analysis to come up with a
better estimate of the number of drug injectors in each city.

It seems intuitively reasonable to multiply the standardized scoring
coefficients by the estimates, add these together, and then divide by
the sum of the standardized scoring coefficients.  I've even seen this
done.  But I haven't seen a good proof that this is correct (it may not
be correct!)

Any advice on how to proceed will be appreciated.

Thanks in advance

Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)