The main obstacle between here and a preliminary result for A_N is the background subtraction. It has been decided that for the time being, I will work on a "traditional" background correction while Steve works to stabilize his method. One sensible first step is to examine the asymmetry as a function of invariant mass. This also serves to satisfy a question raised by Elke at the analysis meeting. She had questioned whether or not the low-mass background is indeed spin-independent and whether or not we had analyzed the asymmetry for m_γγ < 0.1 GeV/c².

Figure 1: A_N vs. m_γγ

Figure 1 shows A_N as a function of invariant mass. One sees that for x_F > 0, the asymmetries everywhere, including below 0.1 GeV/c², are quite small. There appears to be a "bump" around the pion mass, though the effect is quite small statistically speaking. A constant fit to the x_F > 0 points is actually too good: 0.0070±0.0072 with χ²/ν = 2.674/5. A constant fit to the corresponding p₀'s is probably not good enough: 0.0044±0.0051 with χ²/ν = 9.72/5.

More troubling to me is the behavior for x_F < 0. For the bin below 0.1 GeV/c² there is a 3.9σ effect around 7%. A constant fit to the asymmetry points yields 0.0029±0.0071 with χ²/ν = 20.77/5. A similar fit to the p₀ points yields 0.011±0.005 with χ²/ν = 2.751/5. Also, I note that five of six points sit above the horizontal.

Figure 2: Asymmetry Fits in Mass Bins (x_F > 0)

0.0 < mγγ < 0.1 GeV/c2
0.1 < mγγ < 0.2 GeV/c2
0.2 < mγγ < 0.3 GeV/c2
0.3 < mγγ < 0.4 GeV/c2
0.4 < mγγ < 0.5 GeV/c2
0.5 < mγγ < 0.6 GeV/c2

Figure 3: Asymmetry Fits in Mass Bins (x_F < 0)

0.0 < mγγ < 0.1 GeV/c2
0.1 < mγγ < 0.2 GeV/c2
0.2 < mγγ < 0.3 GeV/c2
0.3 < mγγ < 0.4 GeV/c2
0.4 < mγγ < 0.5 GeV/c2
0.5 < mγγ < 0.6 GeV/c2