<div dir="ltr"><br><div class="gmail_quote">---------- Forwarded message ----------<br>From: <b class="gmail_sendername">Dave Mack</b> <span dir="ltr"><<a href="mailto:mack@jlab.org">mack@jlab.org</a>></span><br>Date: Sat, Sep 6, 2014 at 3:43 PM<br>Subject: Re: [Q-weak] First draft of the Qweak elastic transverse asymmetry paper<br>To: Buddhini Waidyawansa <<a href="mailto:buddhini@jlab.org">buddhini@jlab.org</a>><br><br><br>
<div bgcolor="#FFFFFF" text="#000000">
<div>Buddhini,<br>
<br>
So is Table I without polarization correction? If the point of the
table is to convince people that we measured the same result 3
times, then don't we have to divide by 1/P first? You quote an
average polarization so you could easily correct these
experimental asymmetries by 1/<P>. <br>
<br>
However, I assume there were 1-2% polarization differences between
Run I and Run II. Today I've decided to be worried about problems
that might arise from the averaging. Inputting some guesses into a
calculator it looks like the correct average for 3 datapoints<br>
<br>
<Aphys> = ( (Aexp_1/P1)*W1 + (Aexp_2/P2)*W2 + (Aexp_3/P3)*W3
)/(W1+W2+W3)<br>
<br>
will be extremely close to the approximation you apparently used<br>
<br>
<Aphys> = (1/<P>)*(Aexp_1*W1 + Aexp_2*W2 + Aexp_3*W3
)/(W1+W2+W3)<br>
<br>
The difference is less than 0.1%. So we're good. <br>
<br>
I'm printing up your data release talk and your old tech note so I
can remember all this stuff. <br><span class="HOEnZb"><font color="#888888">
<br>
Dave</font></span><div><div class="h5"><br>
</div></div></div></div>
</div><br><br clear="all"><br>-- <br><div class="gmail_signature"><div dir="ltr"><font face="georgia, serif">-------------------------------------------------------------------<br>Buddhini Waidyawansa<br></font>Postdoctoral Fellow<br>
C122,<br><div><font face="georgia, serif">12000 Jefferson Ave,</font></div><div><font face="georgia, serif">Newport News, VA 23602.</font></div><div><font face="georgia, serif">TP 757-912-0410</font></div></div></div>
</div>