[Clas_hadron] Question on Affect of Systematics on Polarization.

[Barry Ritchie]

Phil, if you look more closely at Mike’s figures, it’s not going to be a simple process for Mike to get you a number for the uncertainty. Things may be different because of the skim/PID cuts that appear may have been done on your data (if I understand the start of Chapter 4 of the thesis correctly).  Instantaneous edge position factors into everything. A given energy bin does not have a particular polarization (or polarization uncertainty or polarization correction); this is a channel-by-channel, event-by-event problem based on how far the E-counter bin is from the coherent edge. Then, also, please note, as I mentioned earlier, that perp and para apparently are different at both 1.9 and 2.1 GeV in an E-bin-by-E-bin way, based on how far each bin is at the moment from the instantaneous coherent edge, so the W’s will be messed up from that, too. (See Mike’s slide 10, for example.)

So there’s lots of sources of uncertainty in the phi results that we don’t know here at ASU. Those uncertainties (in addition to the statistical ones you mention), come from the channel-to-channel overall consistency issue, the systematic difference in the polarization for each E-bin based on how far the bin is from the instantaneous edge, plus the para/perp variation, and any skim-related issues. Plus, unfortunately, the 2.1 GeV stuff is just really messy. So I don’t think it’s fair to expect a believable number for the uncertainty without you or Julian doing an analysis like Mike and Volker have done for their channels.

As you say, the number of phi mesons is really small, but you’ve got to deal with the E-bins to get this stuff worked out. Ideally, to work with the new polarization info, you could construct corrected and polarization-weighted E-channel-by-E-channel sets of histograms (which would be a new analysis specific to your channel and could be really tough), and then combine them to get new missing mass histograms, and then use your existing codes to redo the SDME analysis. But you guys could just see if you can focus on getting new weighted values of the polarization by using Mike’s polarization consistency analysis approach as a template for weighting the photon spectrum and then correcting the results in the thesis with those. It’s a big job, either way, but not completely impossible.

---BGR


Professor  Barry G. Ritchie
Department of Physics
Arizona State University
Tempe, AZ  85287-1504

Telephone: (480) 965-4707
Fax: (480) 965-7954

________________________________
From: Philip Cole [mailto:colephil at isu.edu]
Sent: Saturday, January 07, 2012 5:02 PM
To: Michael Dugger
Cc: Barry Ritchie; julian salamanca; Ken Livingston; Franz Klein
Subject: Question on Affect of Systematics on Polarization.

Michael,

The error of 5% on the absolute mean for the 2.1 GeV polarization data was the estimate at the time of the writing of the thesis as I recollect from the g8b meetings.   All these data were shown then; nothing has changed.  However, detailed studies had not been produced such as your work on the polarization using the reaction γ p →p π0. How much do you estimate then that the polarization will vary absolutely from an average of 75% over the interval 1.9 to 2.1 GeV?  For the phi analysis Julian took 200 MeV widths in energy for both the 1.7 to 1.9 GeV and the 1.9 to 2.1 GeV energy regimes.  The bin widths are very broad in energy since there are only 8000 or so phis in total over the two settings.

Thanks,
Phil
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/clas_hadron/attachments/20120108/657aa747/attachment-0001.html