You might be giving caret more credit than is due.  It does not
automatically convert t to p or vice versa.  You have to compute the
desired t/z, based on the p value you want.  For example:

If you want to show a metric thresholded at p=.01, and you are feeding
caret t values (df=17), then you'd need to enter 2.8982 in the Caret user
threshold setting for Caret to suppress values below that level.

Similarly, if you feed it z values, you'd enter 2.5758 (corresponding to
p=.01).

The user scale adjusts how the palette maxes/mins out.  The threshold just
makes anything below the threshold transparent, so that any underlay shows
through those vertices.


> <html><head></head><body><div style="font-family: Verdana;font-size:
> 12.0px;"><div>
> <div style="font-family: Verdana; font-size: 12px; line-height:
> normal;"><span style="font-family: verdana, geneva, sans-serif;">Hello
> all,&nbsp;</span></div>
>
> <div style="font-family: Verdana; font-size: 12px; line-height:
> normal;"><span style="font-family: verdana, geneva, sans-serif;"><span
> style="line-height: 1.6em;">I&#39;ve done two&nbsp;metaanalyses (using the
> MKDA approach) and mapped the resulting volumes to a</span><span
> style="line-height: 1.6em;">&nbsp;surface in order to obtain&nbsp;the
> overlapping clusters (I created&nbsp;the overlap&nbsp;</span><span
> style="line-height: 1.6em;">via Surface Menu</span><span
> style="line-height: 19px;">-&gt;Region of Interest
> Operations).</span></span></div>
>
> <div style="font-family: Verdana; font-size: 12px; line-height:
> normal;"><span style="font-family: verdana, geneva, sans-serif;"><span
> style="line-height: 19px;">The MKDA-outputs, which I mapped to the surface
> are pre-thresholded and contain a more liberal threshold than I want to
> have for my data. I know that there is a &quot;Threshold adjustment&quot;
> as well as a &quot;Color Mapping&quot; option in the metric settings menu.
> But I&#39;m not quite sure about the difference. Anyway, my main problem
> is that the MKDA output just provides me with z-values. Is there a
> possibility to threshold my metrics with these z-values (instead of
> t-values, what I suppose is to be done in metric settings/threshold
> adjustments)?</span></span></div>
>
> <div style="font-family: Verdana; font-size: 12px; line-height:
> normal;"><span style="font-family: verdana, geneva, sans-serif;"><span
> style="line-height: 19px;">Lots of thanks in advance</span></span></div>
>
> <div style="font-family: Verdana; font-size: 12px; line-height:
> normal;"><span style="font-family: arial, helvetica, sans-serif;"><span
> style="line-height: 19px;"><span style="font-family: verdana, geneva,
> sans-serif;">Fabian</span></span></span></div>
> </div></div></body></html>
> _______________________________________________
> caret-users mailing list
> [email protected]
> http://brainvis.wustl.edu/mailman/listinfo/caret-users
>


_______________________________________________
caret-users mailing list
[email protected]
http://brainvis.wustl.edu/mailman/listinfo/caret-users

Reply via email to