Strong evidence of test–retest reliability (ICC: 2- to 5-years’ population = 0.94–0.95 2- to 25-years’ population = 0.97), internal consistency (Cronbach’s α: 2- to 5-years’ population = 0.96 2- to 25-years’ population = 0.95), convergent validity (2- to 5-years’ population: CGI-S rho = − 0.84, Vignos grade rho = − 0.79 2- to 25-years’ population: CGI-S rho = − 0.49), and known-groups validity (all P < 0.001) were demonstrated. A total of 165 individuals were included in the analyses, of whom 84 were in the NMD group (aged 2–5 years) and 81 were in the SMA group (aged 2–25 years). The analysis was performed on a dataset provided by Hospices Civils De Lyon, extracted from the multinational MFM32 database. Test–retest reliability (intraclass correlation coefficient ), internal consistency (Cronbach’s alpha ), convergent validity (Spearman rank-order correlations), and known-groups validity (analysis of covariance comparing groups defined by the Clinical Global Impression of Severity scale and Vignos grade) were calculated. To investigate the validity and reliability of the 32-item Motor Function Measure (MFM32) in individuals with neuromuscular disorders (NMD), including spinal muscular atrophy (SMA), aged 2–5 years, and in non-ambulant individuals with Types 2 or 3 SMA, aged 2–25 years.