just to be totally obsessive,  here's the link and the heart of the
statistical validation. Note: p=0.011, which based on some semi educated
googling, indicates, according to the pretty little chart here:
http://en.wikipedia.org/wiki/Standard_normal_distribution#Standard_deviation_and_confidence_intervals

that this result is three standard deviations from normal, in other words
very unlikely to have occurred  by chance.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3092984/

The gray matter volumes of ACC and the right amygdala allowed the
classifier to distinguish individuals who reported themselves as
conservative from those who reported themselves as very liberal with a high
accuracy (71.6% ± 4.8% correct, p = 0.011). This suggests that it is
possible to determine the self-expressed political attitude of individuals,
at least for the self-report measure we used, based on structural MRI scans.



now the rest of the number crunching, : matter volume of anterior cingulate
cortex and right amygdala from each individual to train a multivariate
classifier [9 <http://www.ncbi.nlm.nih.gov/pubmed/16791142>]. A
leave-one-out procedure with cross-validation was used to determine how
well this classifier could predict whether an individual was conservative
or very liberal when trained on the other participants' data

you got me. No clue.

On Wed, Feb 15, 2012 at 5:08 PM, Dana <dana.tier...@gmail.com> wrote:

> pretty sure that's right
>
>
> On Wed, Feb 15, 2012 at 5:03 PM, Maureen <mamamaur...@gmail.com> wrote:
>
>>
>> Depends on the standard deviation.  You can draw valid conclusions for
>> n < 100 if standard deviation < 2.
>>
>> On Wed, Feb 15, 2012 at 4:54 PM, LRS Scout <lrssc...@gmail.com> wrote:
>> >
>> > The sampling of 90 people is really really small.
>>
>> 

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~|
Order the Adobe Coldfusion Anthology now!
http://www.amazon.com/Adobe-Coldfusion-Anthology/dp/1430272155/?tag=houseoffusion
Archive: 
http://www.houseoffusion.com/groups/cf-community/message.cfm/messageid:346957
Subscription: http://www.houseoffusion.com/groups/cf-community/subscribe.cfm
Unsubscribe: http://www.houseoffusion.com/groups/cf-community/unsubscribe.cfm

Reply via email to