deletions | additions       diff --git a/We_are_assuming_that_the__.tex b/We_are_assuming_that_the__.tex  index 97ce850..86a4e52 100644  --- a/We_are_assuming_that_the__.tex  +++ b/We_are_assuming_that_the__.tex 
...
Caramanos 2010 looked specifically at repositioning in relation to SIENA, which is longitudinal analysis, but one could assume that repositioning affected the scan-rescan variability of some ROIs in our cross-sectional analysis.  We are assuming that the consistency of positioning varies between sites, because we did not define a standardized positioning protocol. A big reason for this is because we want to be able to incorporate data that was already acquired for future studies. In the scaling factor derivation, we take the average of the two volumes to calibrate, and calculate an "overall" bias, but we cannot say much about the bias due just to positioning. However, we do have some measure of positioning consistency that is calculated via the amount of variability due to "run" in our ICC calculation - variability in positioning will certainly contribute to the variability between runs. This variance component is much smaller than the variance due to subjects/sites/subjects x site/unexplained, so we will \textbf{report this in a supplemental table}