The performance of optical refractive index sensors is described by the sensor sensitivity and the detection limit that is a function of factors such as the resolution of the output spectrum and the data processing methods. In sensors that use low quality factor resonators or conventional spectrometer, choosing appropriate data processing methods can improve sensor performance. In this paper, the value of the noise variance of the sensor output spectrum is evaluated for curve fitting and Fourier transform methods and it is shown that the Fourier transform method is not as much affected by spectral noises and can give a smaller variance in the output spectrum. The obtained results are important in designing and analyzing the data of these systems.