A Chinese version of the Neonatal Neurobehavioral Examination (NNE-C) was applied to 15 high-risk infants and five normal term infants for investigation of reliability. The infants were assessed by three physical therapists to examine inter-rater reliability and reassessed by one of the therapists within 2 days to examine test-retest reproducibility. The internal consistency of the NNE-C scale was high, with an alpha coefficient of 0.84. The inter-rater reliability was high for item scores (kappa coefficients > 0.75 for 81% of the items) and for section and total scores (all intraclass correlation coefficients > 0.80). The test-retest reproducibility was moderate for item scores (kappa coefficients > 0.40 for 85% of the items) and was high for section and total scores (all intraclass correlation coefficients > 0.80). We conclude that the NNE-C scale is clinically feasible and reliable for the evaluation of neurobehavioral functions of high-risk and normal term infants in Chinese-speaking societies.