1: \pagestyle{plain}
2: \section{Analysis Introduction} \label{qeff_intro}
3:
4: In this chapter a description of the event reconstruction procedure
5: and the corrections and efficiencies applied to the measured beam charge, $Q$, will be presented.
6: These corrections and efficiencies include the computer and electronics
7: livetimes , $CLT$, and $ELT$, respectively, VDCs tracking efficiency, $\epsilon_{VDC}$, VDCs hardware cuts
8: efficiency $\epsilon_{VDCH}$, scintillators efficiency (product of the two scintillators efficiency
9: or $\epsilon_{S_1} \times \epsilon_{S_2}$), particle identification (PID) efficiency,
10: $\epsilon_{PID}$, proton absorption correction, $C_{Absorption}$, target boiling
11: correction, $C_{TB}$, and target length correction, $C_{TL}$.
12: In addition, a brief description of the spectrometer optics calibration and spectrometer
13: mispointing measurements will be given.
14: Having applied these corrections and efficiencies to the beam
15: charge, we refer to the corrected charge as the effective charge or $Q_{eff}$ and
16: it is defined as:
17: %
18: %%%\begin{equation}
19: \begin{eqnarray} \label{eq:qeff}
20: Q_{eff} = \frac{1}{ps} \Big(Q \times ELT \times CLT \times \epsilon_{VDC} \times \epsilon_{VDCH}
21: \times \epsilon_{S_1} \times \epsilon_{S_2} \times \epsilon_{PID}
22: \times C_{Absorption} {}
23: \nonumber\\
24: \times C_{TB} \times C_{TL}\Big)~,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{}
25: \end{eqnarray}
26: %%%\end{equation}
27: %
28: where $ps$ is the prescale factor. The prescale factor $n$ for the trigger type T$_i$
29: ($i = 1,\cdots,5$) means that the Trigger Supervisor will read out every $n$th event of
30: type T$_i$.
31:
32: The advantages of detecting protons as against electrons were discussed in section \ref{proton_vs_electron}.
33: Because of these advantages, the E01-001 experiment was able to greatly reduce any
34: $\varepsilon$-dependent systematic corrections and associated uncertainties applied to the measured cross
35: sections. In addition, measurements at $Q^2$ = 0.5 GeV$^2$ using the right arm spectrometer which
36: served as a luminosity monitor will check the uncertainties due to beam charge, current, and target
37: density fluctuations.
38:
39: %%%%%%%%%%%%%%%%%%%%%%
40: \pagestyle{myheadings}
41: %%%%%%%%%%%%%%%%%%%%%
42:
43: The systematic uncertainties in these efficiencies and corrections will be discussed.
44: These systematic uncertainties are a measure of how accurately we know the various efficiencies and
45: corrections. The systematic uncertainties are classified into three types, based on how they contribute
46: to the quantities we want to extract:
47: \begin{itemize}
48: \item {Scale Uncertainty: Sometimes referred to as normalization uncertainty as
49: it has the same effect on all $\varepsilon$ points at a given Q$^2$. An example would be the
50: uncertainty in the target length. Such uncertainty would affect both $G_{Ep}$ and $G_{Mp}$ the
51: same way but not the ratio.}
52: \item {Random Uncertainty: Sometimes referred to as point-to-point uncertainty. It has
53: an uncorrelated effect on each $\varepsilon$ point at a given Q$^2$. Two examples are
54: the statistical uncertainty and the uncorrelated shifts in the beam energy and scattering angle for different
55: kinematics. Such uncertainties affect the extraction of $G_{Ep}$, $G_{Mp}$, and the ratio.}
56: \item {Slope Uncertainty: Refers to a correlated uncertainty in a correction that varies linearly
57: with $\varepsilon$. An example would be the effect of a fixed scattering angle or
58: beam energy offset for all kinematics which also leads to a scale uncertainty. Such an uncertainty
59: will modify the slope of the reduced cross section verses $\varepsilon$, and thus $G_{Ep}$ and $G_{Ep} \over G_{Mp}$, but will not
60: spoil the expected linearity.}
61: \end{itemize}
62:
63: It must be mentioned that in some cases it is unclear whether the uncertainty is scale, random, or
64: slope. In these cases, the worst case of the three uncertainties is assumed.
65: Table \ref{uncert_contrib} indicates whether the different types of uncertainties effect
66: the individual form factors, their ratio, and the linearity of the L-T plots. Note that the slope uncertainty is
67: the change in slope between $\varepsilon$ = 0.0 and $\varepsilon$ = 1.0. So the value for the slope uncertainty is the
68: change over the actual $\varepsilon$ range in data divided by $\Delta \varepsilon$.
69: Since $\Delta \varepsilon$ $\sim$ 0.70 for the left arm and $\Delta \varepsilon$ $\sim$ 0.07 for the right arm, the
70: slope uncertainty is much larger for the right arm.
71:
72: %Such uncertainty affects the linearity of the L-T plots since any deviation from linearity
73: %would provide a clear signature of the two-photon-exchange (TPE), and would provide information on the
74: %nonlinear component of the TPE.}
75:
76: \begin{table}[!htbp]
77: \begin{center}
78: \begin{tabular}{||c|c|c|c|c||} \hline \hline
79: Uncertainty Type & $ G_{Ep}$ & $G_{Mp}$ & $\mu_{p}G_{Ep}$/$G_{Mp}$ & Linearity \\
80: \hline \hline
81: Scale & Yes & Yes & No & No \\
82: Random & Yes & Yes & Yes & Yes \\
83: Slope & Yes & Yes & Yes & No \\
84: \hline \hline
85: \end{tabular}
86: \caption[The effect of the scale, random, and scale uncertainties on the individual form
87: factors, their ratio, and the linearity of the L-T plots.]
88: {The effect of the scale, random, and scale uncertainties on the individual form
89: factors, their ratio, and the linearity of the L-T plots.}
90: \label{uncert_contrib}
91: \end{center}
92: \end{table}
93: