[d2n-analysis-talk] cos(phi) in two 4.7-GeV run sets

[Diana Parno]

Hello all,

After our meeting today, I finished running comparison code for the  
cosine of the azimuthal angle (phi in terms of the physics,  
BB.tr.tg_th in terms of analyzer variables). (Recall that all  
kinematic variables not related to phi were consistent with each other  
in all x bins.) You can see the result below. Red points are for the  
earlier data set, Runs 2024-2035 (which Matt had been using as a  
metric) and blue points are for the later data set, Runs 2150-2163  
(which I had been using as a metric). Matt's run set has somewhat more  
events than mine.

The two points with the worst agreement are in the zeroth bin  
(x=0.175) and in the fourteenth bin (x = 0.875). At x = 0.175, the  
values of cos(phi) differ by 2.8 sigma. At x = 0.875, they differ by  
3.2 sigma.

However, cos(phi) enters into the asymmetries as a coefficient, and if  
you look at the y-axis scale you see that we are talking about very,  
very small variations. At x = 0.175, the blue point is 0.011% higher  
than the red point; at x = 0.875, it is 0.043% higher. (Yes, we are  
talking parts in ten thousand.) This is a negligible source of error  
on the asymmetry. I think it would be worthwhile to repeat this check  
after the five-pass data are analyzed, sampling a few different run  
sets to ensure that there isn't a big shift between any of them, but  
we should not lose any sleep on this.

Best,
Diana

-------------- next part --------------
A non-text attachment was scrubbed...
Name: mean_cosphi.pdf
Type: application/pdf
Size: 14488 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20110824/01991f1f/attachment.pdf 
-------------- next part --------------