Parallel plate geometry at narrow gaps allows one to perform rheology at high shear rates and with small volumes of fluid. The results from this analysis have to be corrected for a systematic error in the actual gap between the plates. We review the statistical and error analysis methodology for such applications and demonstrate that the process for estimating the gap error significantly impacts the overall accuracy of the analysis at narrow gaps. It is shown that the standard linear regression approach to the estimation of the gap correction is unsound and must not be used in practice. We highlight alternative methods and illustrate their accuracy with typical examples from narrow gap rheology. (C) 2013 The Society of Rheology. [http://dx.doi.org/10.1122/1.4774323]