This study utilises moisture distribution profiles harvested from literature and gravimetric observations made in the laboratory to analyse the robustness of three Boltzmann-profile, ?n(?) models which were previously adopted by various researchers for the modelling of non-linear hydraulic diffusivity, D(?n) of drying mortar and concrete. Such an analysis is necessary to ensure that the D(?n) data is reliable and free from modelling artefacts. A 2-parameter ?n(?) model is found to be better suited for the inverse estimation of D(?n) in comparison to a widely used 3-parameter model. The 2-parameter model captures the effects of water-to-cement ratio and temperature on D(?n) in accordance to the physics of the drying phenomenon. Also, it does not cause a sudden drop in the diffusivity curve near high saturation levels as is noted with the 3-parameter model. It is further noted that the 3-parameter ?n(?) model fails to fit low ?n - low ? data which is typically obtained using non-destructive imaging techniques. The 2-parameter model can fit low ?n - low ? data and is also suitable for high ?n - low ? data, obtained through gravimetric experiments, within the limit of ?n < 0.80.