Gait disorders can be disabling in persons with multiple sclerosis (PMS). Different gait parameters have been used to evaluate gait disorders according to the International classification of functioning. Some authors have reported a direct relation between evaluations over short distances and long-term outcomes. This relationship is of interest for the purposes of clinical research, as it enables short-distance evaluations to be used as a primary endpoint for trials. However, these endpoints are not always particularly relevant for PMS, and furthermore, all evaluations do not present the same metrological characteristics, especially with regards to reproducibility. However, it is essential to have good reproducibility in order to be able to test the effect of a therapeutic strategy on walking parameters in PMS. Using a range of walk tests (timed 25-foot walk in different conditions, namely comfortable walking, fast walking and dual-task walking; the Timed Up and Go test; the 6minute walk test) associated with neuromotor analysis of the lower limbs, we describe the advantages and limitations of gait evaluation in MS. Based on clinical experience accumulated over 4 years, we propose a minimum set of measurements to be used in clinical practice and also for research purposes.