[RkWard-devel] Pareto chart and Quality qontrol charts

I. Soumpasis nono.231 at gmail.com
Fri Feb 23 21:34:16 UTC 2007


2007/2/23, Thomas Friedrichsmeier <thomas.friedrichsmeier at ruhr-uni-bochum.de
>:
>
> On Friday 23 February 2007 21:21, I. Soumpasis wrote:
> > I disagree here because of the meaning of the data. Pareto chart is used
> > when you have data like this
> > A B C D E F
> > 1 1 1 2 2 3
> > Suppose that the above is type of defects, you can take the pareto chart
> > with the most serious of them. It has a different approach of the
> barplot.
> > The summary(factor) gives the the factor variable in this way to help
> the
> > user not to have to make a new variable in order to use the plugin.
>
> True, but summary (factor) is essentially identical to table (factor).
> (More
> precisely: table (factor, exclude=NULL) to keep the NAs, so that would be
> more correct).
>
> > But the pareto.chart (and the plugin) is meaningless for
> > a <- rep (c (1, 1, 1, 2, 2, 3), length.out=100)
>
> But what if 1, 2 and 3 are actually three different types of failures (and
> the
> user simply didn't define this as a factor)? Or a count of defects on a
> single unit (perhaps pixel errors on an LCD, and the defect count
> influences
> the sale category), so it's actually reasonable to store in a vector
> instead
> of a factor? In these cases, table sums them up nicely, as:
>
> > table (a)
> a
> 1  2  3
> 51 33 16
>
> > If these are the names the user must make them factor  so to make a
> > pareto.chart() and use the plugin.
>
> The question I'm having is this:
> Are there any vectors (not factors, and not tables) that could
> meaningfully be
> displayed as raw data? If not, then either the plugin should not accept
> vectors at all, or it should convert them to something that it can display
> in
> a meaningful way. If there are, then it may still be worth considering,
> whether the type of data I suggested should not be supported in some way
> as
> well, perhaps with a "tabulate" option.


I used the table option as you proposed. I do not know all the probable uses
of pareto.chart, but the most common, so I cannot say anything for sure.

Regards,
Ilias
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/rkward-devel/attachments/20070223/b75fc754/attachment.html>


More information about the Rkward-devel mailing list