The validity and reliability of the guidelines of the Surveillance of Cerebral Palsy in Europe (SCPE) for the classification of cerebral palsy (CP) were tested by administering 10 written case vignettes via an interactive web-based link to 30 SCPE partners. There was a moderately good level of agreement (kappa=0.59) about inclusion as a CP case on the SCPE database. Classification by CP subtype differed in two main areas: assigning spastic versus dyskinetic and judgement of distribution of spastic involvement. Agreement on Gross Motor Function Classification System (GMFCS) level was less good than reported in previous studies. Twenty respondents repeated the test 5 months later and there was good repeatability for case inclusion (kappa= 0.72) but considerable variation in assignment of CP subtype and GMFCS level. There is a need for further collaborative work and training to improve harmonization of the classification of CP, including examination, application of SCPE guidelines, and register coding.