[rkward-devel] New plugin and functions

[I. Soumpasis]

2007/4/19, Thomas Friedrichsmeier <thomas.friedrichsmeier at ruhr-uni-bochum.de
>:
>
> Hi Ilias,
>
> On Saturday 14 April 2007 10:42, I. Soumpasis wrote:
> > I uploaded a new plugin, the crosstabs plugin. I wait for
> recommendations,
> > corrections etc.
>
> ok, I had a first look at the plugin. Certainly a rather important one.
>
> Some rather general comments:
> 1) The barplot functionality should be made an embedded plugin so the code
> can
> be shared between the standalone barplot and this one. If you don't know
> how
> to approach this, I can take a look, later, but if you have the time to
> look
> at this, that would certainly be good.


That was my thought in the first place but did not know how to do it and I
was in great harry to get results for the data I had.

2) Same actually for the chisquare test, except we don't have this as a
> standalone plugin, yet.


I agree on this too but still do not know how to do it.

3) Perhaps the most basic: table() will happily take any number of
> arguments.
> Barplot and chisquare will only work on two-dimensional tables, but the
> crosstabulation itself is not limited in this way. We could make it so
> that
> the chisquare and barplot options are only available if there are no more
> than two objects selected (I'll need to add some simple C code to make
> this
> possible). Would this sound like a good solution?


Here are my thoughts. (Limited responsibility, because there is some time
passed since I wrote it nd I do not remember all the things I thought then
plus I do not have an SPSS right now).
1. A small problem we have to see. If we use more than two variables the
format of the result changes. As I remember of the SPSS crosstabulation they
have one dependent and many independent variables. In this way we could
crosstabulate the dependent with each independent and then put them in a
table. Do you find this a sound solution?
2. SPSS also does one barplot for every crosstabulation. With the above
solution is easiest to be done for ever pair dependent-independent we have.
The only minus is that we have to stop offering preview, something  logical
because here we have an analysis, not a plotting plugin.
3. Same with chisquare test.

4) Some of the code in the PHP file should be moved to the calculate
> section.
>
> 5) You can directly use the xy object also for the chisq.test.
>
> 6) Don't be afraid to create temporary variables any more. Esp. for the
> chisq.test(), it is not necessary to actually calculate the test three
> times.
> I suggest rewriting this to:
>
> # in calculate()
> chisqres <- chisq (xy)
>
> # in printout()
> rk.results (list
> ('Statistic'=chisqres$statistic, 'df'=chisqres$parameter,
> 'p-value'=chisqres$p.value))


Yes you are right, but I will wait for what we are finally going to do with
the whole plugin.


Regards,
Ilias
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/rkward-devel/attachments/20070419/bd27d0fe/attachment.html>