Login  Register

Correlated factors or not.

classic Classic list List threaded Threaded
2 messages Options Options
Embed post
Permalink
Reply | Threaded
Open this post in threaded view
| More
Print post
Permalink

Correlated factors or not.

Mark Webb-3
65 posts
I've used .....
      Extraction Method: Principal Component Analysis.
       Rotation Method: Varimax with Kaiser Normalization.

to extract 16 factors [1 Eiganvalue approach used].
In theory these factors should be uncorrelated. When I do Pearson Correlation I get the following.
There seems to be correlation [e.g. 0.78] - why is this ?

        Fac1  
      Factors Pearson Correlation Sig. (2-tailed) N
      Fac1 1.00  547.00
      Fac2 0.78 0.00 547
      Fac3 0.52 0.00 547
      Fac4 0.63 0.00 547
      Fac5 0.51 0.00 547
      Fac6 0.37 0.00 547
      Fac7 0.48 0.00 547
      Fac8 0.46 0.00 547
      Fac9 0.63 0.00 547
      Fac10 0.43 0.00 547
      Fac11 0.60 0.00 547
      Fac12 0.61 0.00 547
      Fac13 0.61 0.00 547
      Fac14 0.58 0.00 547
      Fac15 -0.16 0.00 547
      Fac16 0.29 0.00 547
Reply | Threaded
Open this post in threaded view
| More
Print post
Permalink

Re: Correlated factors or not.

statisticsdoc
249 posts
Stephen Brand
www.statisticsdoc.com

Mark,

This is quite odd.  I have know of instances in which the saved factor scores from an orthgonal rotation were weakly correlated (because the factor scores are estimated), but the correlations that you report is very large.

How did you obtain this correlation matrix?  Did you calculate and save factors scores?  If so, check the syntax for the calculation.  If not, could you post or send me the syntax that you utilized in this analysis?

HTH,

Steve


---- Mark Webb <[hidden email]> wrote:

> I've used .....
>       Extraction Method: Principal Component Analysis.
>        Rotation Method: Varimax with Kaiser Normalization.
>
> to extract 16 factors [1 Eiganvalue approach used].
> In theory these factors should be uncorrelated. When I do Pearson Correlation I get the following.
> There seems to be correlation [e.g. 0.78] - why is this ?
>
>         Fac1
>       Factors Pearson Correlation Sig. (2-tailed) N
>       Fac1 1.00  547.00
>       Fac2 0.78 0.00 547
>       Fac3 0.52 0.00 547
>       Fac4 0.63 0.00 547
>       Fac5 0.51 0.00 547
>       Fac6 0.37 0.00 547
>       Fac7 0.48 0.00 547
>       Fac8 0.46 0.00 547
>       Fac9 0.63 0.00 547
>       Fac10 0.43 0.00 547
>       Fac11 0.60 0.00 547
>       Fac12 0.61 0.00 547
>       Fac13 0.61 0.00 547
>       Fac14 0.58 0.00 547
>       Fac15 -0.16 0.00 547
>       Fac16 0.29 0.00 547

--
For personalized and experienced consulting in statistics and research design, visit www.statisticsdoc.com