The conduct and interpretation of electromyography in children is considered difficult and therefore often avoided. We assessed the diagnostic accuracy of the paediatric electromyography protocol used in our tertiary reference centre and compared it to muscle biopsy results and clinical diagnosis. Electromyography was performed in unsedated children with suspected neuromuscular diseases between January 2010 and September 2017 and was analysed quantitatively. Muscle pathology was classified into seven groups based on existing histopathology reports. The clinical diagnosis, including myopathic, neurogenic and non-neuromuscular categories was used as the gold standard. 171 children between the age of 12 days to 17.4 years were included in the analysis. 41 children (24%) were under the age of 2 years at the time of electromyography. 98 (57%) children were diagnosed with a myopathic disorder, 18 (11%) with a neurogenic disease and 55 (32%) as not having a primary neuromuscular disorder. In detecting myopathic disease, electromyography performed as well as muscle biopsy (sensitivity 87.8% for electromyography vs. 84.5% for muscle biopsy; specificity 75.7% vs. 86.4%). This also applied to children under the age of 2 years (sensitivity 81.8% vs. 86.4%). Quantitative analysis of a limited electromyography protocol performed in unsedated children is a very valuable diagnostic tool.