[R] package ‘contingency.tables’ is not available (for R version 2.15.2)

[Milan Bouchet-Valat]

Le lundi 04 mars 2013 à 14:43 -0600, Joanna Papakonstantinou a écrit :
> 
> Thank you. 
> I actually ended up using:
> > CrossTable(mdt)
> 
>  
>    Cell Contents
> |-------------------------|
> |                       N |
> | Chi-square contribution |
> |           N / Row Total |
> |           N / Col Total |
> |         N / Table Total |
> |-------------------------|
> 
>  
> Total Observations in Table:  8 
> 
>  
>              |  
>              |      Blue |     Green |       Red | Row Total | 
> -------------|-----------|-----------|-----------|-----------|
>            A |         1 |         0 |         1 |         2 | 
>              |     0.083 |     0.500 |     0.083 |           | 
>              |     0.500 |     0.000 |     0.500 |     0.250 | 
>              |     0.333 |     0.000 |     0.333 |           | 
>              |     0.125 |     0.000 |     0.125 |           | 
> -------------|-----------|-----------|-----------|-----------|
>            G |         1 |         0 |         1 |         2 | 
>              |     0.083 |     0.500 |     0.083 |           | 
>              |     0.500 |     0.000 |     0.500 |     0.250 | 
>              |     0.333 |     0.000 |     0.333 |           | 
>              |     0.125 |     0.000 |     0.125 |           | 
> -------------|-----------|-----------|-----------|-----------|
>            H |         1 |         0 |         1 |         2 | 
>              |     0.083 |     0.500 |     0.083 |           | 
>              |     0.500 |     0.000 |     0.500 |     0.250 | 
>              |     0.333 |     0.000 |     0.333 |           | 
>              |     0.125 |     0.000 |     0.125 |           | 
> -------------|-----------|-----------|-----------|-----------|
>            W |         0 |         2 |         0 |         2 | 
>              |     0.750 |     4.500 |     0.750 |           | 
>              |     0.000 |     1.000 |     0.000 |     0.250 | 
>              |     0.000 |     1.000 |     0.000 |           | 
>              |     0.000 |     0.250 |     0.000 |           | 
> -------------|-----------|-----------|-----------|-----------|
> Column Total |         3 |         2 |         3 |         8 | 
>              |     0.375 |     0.250 |     0.375 |           | 
> -------------|-----------|-----------|-----------|-----------|
> 
> and 
> > expected.counts <- (apply(mdt,1,sum) %o% apply(mdt,2,sum))/sum(mdt)
> 
> 
> > print(expected.counts)
> 
> 
>   Blue Green  Red


> A 0.75   0.5 0.75


> G 0.75   0.5 0.75


> H 0.75   0.5 0.75


> W 0.75   0.5 0.75
> 
> 
> to get the info I needed.
>  
Actually, CrossTable(..., expected=TRUE) should give you the same
result.


Regards

> Regards,
> Joanna
> 
> 
> On Mon, Mar 4, 2013 at 2:15 PM, Milan Bouchet-Valat
> <nalimilan at club.fr> wrote:
>         Le lundi 04 mars 2013 à 12:37 -0600, Joanna Papakonstantinou a
>         écrit :
>         > I am trying to create contingency tables (to evaluate prior
>         to performing
>         > Pearson's Chi-Squared test for independence). I would like
>         to see column
>         > and row totals as well as expected and observed values and
>         cell counts.
>         > I tried to use the package "contingency. tables" but get the
>         following
>         > warning:
>         >  package contingency.tables is not available (for R versions
>         2.15.2)
>         > Is there something comparable or was this replaced with
>         something else for
>         > later versions of R.
>         > Any help would be appreciated so that I may create a
>         contigencey table like
>         > this.
>         
>         I'm not aware of the existence of such a package, but base R
>         provides
>         most of what you need, and other packages offer even more. See
>         e.g.
>         http://www.statmethods.net/stats/frequencies.html
>         
>         
>         Regards
> 
> 
> 
> -- 
> ******************************************************************
>  
> Joanna Papakonstantinou, Ph.D.
> 
>