The determination of strain from neutron diffraction data is normally based upon the fit of a Gaussian function to a Bragg reflection. The error in the fit is assumed to be that based on ‘counting statistics’ and this error propagates through the analyses until the final stress evaluation. This relies on there being a big enough number of diffracting grains/crystallites within the gauge volume to ‘approximate’ to counting statistics. The number of grains however depends on the gauge volume size chosen and the average size of the grains (and hence diffracting grains) within the gauge volume and this should be taken into account. The aim of this work is to give an estimate of the uncertainty due to these ‘grain-size statistics’ due to grain size, gauge volume, FWHM of the Bragg reflection (for angular dispersive diffractometers), scattering angle (2), size of detector (and hence number of diffracting grains ‘seen’ on the detector), hkl multiplicity (m) and eventually texture.