Hi Gavin,
I can't work through this right now, but I would start by looking at
the 'hetcor' package to get the correlations, or if they are already
in the return object, build a plot from these.
Jim
On Wed, Feb 22, 2023 at 4:24 AM gavin duley <gduley at gmail.com>
wrote:>
> Hi!
>
> Apologies if this is not the correct place to ask. I am attempting a
> MFA analysis of a dataset based on wine chemical and sensory analysis,
> based on the STHDA tutorial [1]. (I am using this dataset here too, as
> an example dataset to work on without posting my actual data. I've
> tried this with both my data and the example data, with the exact same
> results.)
>
> The only issue I am having is that I would like to produce a graph
> showing the correlation between qualitative variables, quantitative
> variables, and dimensions for some but not all analyses types. For
> example, it would be good to see only origin, odor, and
> odor.after.shaking on the graph.
>
> By default, it doesn't seem possible to include both qualitative and
> qualitative data in the correlation graph, or to filter by
> res.mfa$separate.analyses.
>
> I am using the code:
> fviz_mfa_var(res.mfa,
choice=c("quanti.var","group","quali.var"),
> palette = "jco",
> col.var.sup = "violet", repel = TRUE)
>
> I have tried using select.ind with no change. This just produces the
> same graph as above:
> fviz_mfa_var(res.mfa,
choice=c("quanti.var","group","quali.var"),
> palette = "jco",
> col.var.sup = "violet", repel = TRUE,
>
select.ind=list(name=c(res.mfa$separate.analyses$origin,res.mfa$separate.analyses$odor,
> res.mfa$separate.analyses$odor.after.shaking)))
>
> Attempting to simplify by just specifying one name (as a trial) does
> not work either.
>
> Is this actually possible? If so, how would it be best to attempt it?
>
> With thanks,
> gavin,
>
> [1]
http://www.sthda.com/english/articles/31-principal-component-methods-in-r-practical-guide/116-mfa-multiple-factor-analysis-in-r-essentials/
>
> --
> Gavin Duley <gduley at gmail.com>
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.