The evaporation of small droplets of volatile liquids from solid surfaces depends on whether the initial contact angle is larger or less than 90°. In the latter case, for much of the evaporation time the contact radius remains constant and the contact angle decreases. At equilibrium, the smaller the drop, the more it is possible to neglect gravity and the more the profile is expected to conform to a spherical cap shape. Recently published work suggests that a singular flow progressively develops within the drop during evaporation. This flow might create a pressure gradient and so result in more flattening of the profile as the drop size reduces, in contradiction to expectations based on equilibrium ideas. In either case, it is important to develop methods to quantify confidence in a deduction of elliptical deviations from optically recorded droplet profiles. This paper discusses such methods and illustrates the difficulties that can arise when the drop size changes, but the absolute resolution of the system is fixed. In particular, the difference between local variables, such as contact angle, cap height, and contact diameter, which depend on the precise location of the supporting surface, and global variables such as radii of curvature and eccentricity, is emphasized. The applicability of the ideas developed is not limited to evaporation experiments, but is also relevant to experiments on contact angle variation with drop volume.