[rkward-devel] power analysis plugin

meik michalke meik.michalke at uni-duesseldorf.de
Sun Oct 5 10:00:05 UTC 2014


hi,

Am Samstag, 4. Oktober 2014, 12:08:49 schrieb meik michalke:
> Am Samstag, 4. Oktober 2014, 10:34:58 schrieb Thomas Friedrichsmeier:
> > - Don't forget to require(pwr)

fixed.

> > I figured out that for "less" you want to specify a _negative_ effect size
> > (and in fact, you can't enter that in the GUI).
> 
> good catch. yes, right now effect sizes can't be negative, i'll try to come
> up with something.

effect size range now covers -1 to 1 globally. it's pretty usual for 
correlations, at least.

> > I suggest, either going with "greater", only, or showing a warning, when
> > the specified effect size contradicts the hypothesis.
> 
> if i take the first road and go astray from the pwr options, i'll probably
> rename "greater" to something like "one sided test (first is greater)".

i included a warning instead. it's also shown if you give a negative effect 
size for the other alternatives, respectively.

> > - I wouldn't write the results to the workspace _by default_.

fixed.

i also fixed the sample size controls for two sample designs, and added the 
possibility to provide eta squared instead of cohen's f.

to make it easier to test, i've released 0.01-1 last night.


viele grüße :: m.eik

-- 
  dipl. psych. meik michalke
  institut f"ur experimentelle psychologie
  abt. f"ur diagnostik und differentielle psychologie
  heinrich-heine-universit"at d-40204 d"usseldorf
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/rkward-devel/attachments/20141005/a9a0fbdb/attachment.sig>


More information about the Rkward-devel mailing list