[RkWard-devel] new tests

Stefan Roediger stefan_roediger at gmx.de
Wed Feb 28 23:20:48 UTC 2007


Hi list,

tonight I added a bunch of new tests. For those in "TEST" and "variances"  I 
will need a couple of days to go through again to make them ready for your 
review (but you can check also now if you want).The same applies to the 
outlier tests which I sent a couple of days ago. Those in "TEST" need a home 
but since they are different it's not yet easy to say where. We still need to 
discuss the menu layout. Suggestions?
However, the new test should work basically (actually there is no complicated 
stuff from my side). It might be that I forgot to fetch some results and the 
help-files need in general more dedication (i. e. regarding the input factor 
vs. numeric vector ...).

Important to note in these plug-ins is that Thomas introduced two new RKWard 
functions:

* rk.describe.alternative (used in ansari_bradley_test.php, mood_test.php, 
F_test.php)
* rk.print.literal (used in stem-and-leaf plot, right now the only way to get 
a propper output)

They turn out to be very useful so maybe worth for you to keep them in mind. 
Thomas you wrote a lot of such functions. Is there a documentation for them?

On my ToDo list are still:

* oneway.test {stats}
* prop.trend.test {stats}
* prop.test {stats}
* pairwise.prop.test {stats}
* pairwise.t.test {stats}
* friedman.test {stats}
* mantelhaen.test {stats}
* mauchly.test {stats}
* chisq.test {stats}
* mcnemar.test {stats}
* fisher.test {stats}
* anova.glm {stats}
* binom.test {stats}

Is somebody else planing to work on some of these?

Stefan




More information about the Rkward-devel mailing list