Inference of magnetic fields from very high spatial, spectral, and temporal resolution polarized images is critical in understanding the physical processes that form and evolve fine scale structures in the solar atmosphere. Studying high spectral resolution data also helps in understanding the limits of lower resolution spectral data. We compare three different methods for calibrating the line-of-sight component of the magnetic field. Each method is tested for varying degrees of spectral resolution on both synthetic line profiles computed for known magnetic fields and real data. The methods evaluated are: (a) the differences in the center of gravity of the right and left circular components for different spectral resolution, (b) conversion of circular polarization, at particular wavelengths, to magnetic fields using model-dependent numerical solutions to the equations of polarized radiative transfer, and (c) the derivative method using the weak field approximation. Each method is applied to very high spatial and spectral resolution circular polarization images of an active region, acquired in the Fe I 5250 angstrom Zeeman-sensitive spectral line. The images were obtained using the 20 mangstrom pass-band tunable filter at NSO/Sacramento Peak Observatory Vacuum Tower Telescope. We find that the center-of-gravity separation offers the best way of inferring the longitudinal magnetic field.