Ana,

I have read quite a few papers that employ PCA but I have never seen them
refer to the specific loadings values; however, I am primarily reading in
the field chemoinformatics and this may vary in ecological studies (which I
assume this would be given the list).  Generally a loadings plot is
generated alongside the scoresplot.  PC loadings are correlation
coefficients between the PC scores and the original variables.  Therefore
most people simply look for variables clustered in similar areas of the
loadings plot as sample clusters in the scores plot and refer to these
variables with no mention of the specific values.  Depending upon you
particular field this approach may be different but this has been my
experience wherever I have seen PCA applied.

Hope this helps,
Regards,

Charles Determan


On Tue, Apr 1, 2014 at 10:03 AM, Ana Serronha <[email protected]> wrote:

> Hi everyone,
>
> I have read that,  loadings (from pca) more than .5 are typically
> considered strong; loadings between .3 and .5 are acceptable; and loadings
> less than .3 are typically considered weak, am I correct?
> However, I can't find a credible reference to use. Can you suggest me a
> book or a paper to use as a reference?
> Thanks
>
>
>
>
> --
> Ana Marta Serronha
>
> CIBIO, Centro de Investigação em Biodiversidade e Recursos Genéticos
> Campus Agrário de Vairão
> Rua Padre Armando Quintas - Crasto
> 4485-661 Vairão
> Portugal
>
> Tel. +351 252660428 ou +351 252660411
> Fax: +351 252661780
>
>         [[alternative HTML version deleted]]
>
>
> _______________________________________________
> R-sig-ecology mailing list
> [email protected]
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>
>


-- 
Charles Determan
Integrated Biosciences PhD Candidate
University of Minnesota

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-ecology mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

Reply via email to