A method is recommended to evaluate the strength of the relationship between visually-perceived colour differences in a given set of colour pairs and their corresponding predictions made by a colour-difference formula. This method is based on the Standardized Residual Sum of Squares (STRESS) index used in multidimensional scaling, and tests if two colour-difference formulae are or are not statistically significantly different. The same index can be also used to compute intra- and inter-observer variability in visual experiments. Beside important advances made in the field of colour-difference evaluation using reliable visual datasets currently available, the results achieved from the STRESS index indicate that it is not possible to recommend a more uniform colour space with a Euclidean colour-difference formula that is statistically significantly better than CIEDE2000.