Objective: Tourette syndrome is a developmental neuropsychiatric disorder in children, and abnormal corticobasal ganglion connectivity is implied for the pathophysiology. Multiscale entropy, an entropy-based method to measure dynamic complexity at multiple temporal scales, is helpful to disclose the information of brain connectivity. This preliminary study investigated the complexity of resting-state electroencephalogram signals using multiscale entropy in children with Tourette syndrome. Method: Resting-state electroencephalographic (EEG) signals were analyzed by sample entropy and multiscale entropy methods in 10 children with Tourette syndrome and 10 healthy gender- and age-matched controls. Results: Except for the Fp2 channel, the complexity index values in all channels were reduced in children with Tourette syndrome compared with those in normal controls. A statistically significant reduction in EEG complexity was found in the bilateral central, parietal, occipital, and left temporal regions, indicating disturbed brain connectivity in Tourette syndrome. Although there was no difference of complexity in the higher frequency spectra, there was a statistically significant difference of complexity in lower frequency in F3 channel, pointing to the importance of examining a range of time scales in exploring EEG signals. Conclusions: Our preliminary study demonstrated that EEG complexity was significantly lower in children with Tourette syndrome than in normal controls. This difference may serve as a marker of disturbed brain connectivity in such individuals and suggests that further clinical studies are warranted.