Dear spmer
I am studying on the correlation between Full
scale IQ and Cerebral Blood Flow of epilepsy
patients. I have 30 subjects , HMPAO SPECT scan,
1 scan/subject, total of 30 scans. The age of
subjects range from 16 to 44 yeras-old. The Full
scale IQ on WAIS test range from 60 to 120.
I used basic models: simple regression, the
full scale IQ as the only covariate, I got
nothing at uncorrected p=0.01 level. However,
using PET/SPECT models: single-subject:
covariates only, I got several clusters at
uncorrected p=0.01 level, the largest one at
prefrontal region.
My questions are:
(1). What is the difference between PET/SPECT
models single-subject: covariates only and basic
models: simple regression. Is it not allowed to
treat those 30 scans from 30 subjects as scans
from only one subject?
(2). Should I give up this study or can I still
use the PET/SPECT single-subject covariates only
result as a preliminary report, and work on it
further. (i.e. increase the sample size)
(3). Will multiple regression with both age and
full scale IQ make better result?
Any response is appreciated.
Sun, Chen-Ming
VGHTPE Taiwan
Aug. 28, 2001
=====
sunchenming
__________________________________________________
Do You Yahoo!?
Make international calls for as low as $.04/minute with Yahoo! Messenger
http://phonecard.yahoo.com/
|