To study whether the response to tissue culture rabies vaccine in children is dose related, 82 children aged 7-13 years (mean: 9.5 years), weighing 17-40 kg (mean: 25 kg), were divided randomly into three groups. All of the children received purified Vero cell rabies vaccine 2-1-1 intramuscular regimen (two sites on day 0, one site each on days 7 and 21). Group I (n = 29) received a full dose of 0.5 mt (antigenic value: 7.1 international units [IU]/dose), group II (n = 26) received a half dose of 0.25 mL (antigenic value: 3.55 IU/dose), and group III (n = 27) received a one-third dose of 0.15 mL (antigenic value: 2.13 IU/dose). The rapid fluorescent focus inhibition test (RFFIT) was used to measure the neutralizing antibody (Nab). The results showed that all of the children seroconverted (> 0.5 IU/mL) on day 14. The Nab responses were dose related. The geometric mean titers (GMT) on days 14, 28, 91, and 182 of group I were 23 (range: 4-134), 18 (range: 3-86), 3 (range: 1-25), and 1 (range. < 0.5-18) IU/mL, respectively. The GMT values of group II on the corresponding days were 20 (range: 5-186), 11 (range: 3-159), 1 (range: 0.5-14), and 1 (range: < 0.5-9) IU/mL. Although the Nab levels in group II were lower than those in group I, the differences were not statistically significant. The GMT of group III on days 14 and 28, which were 6 (range: 0.5-29) and 7 (range: 1-29) IU/mL, were significantly lower than those of groups I and II (P < 0.05). However, the GMT on days 91 and 182 of group III, 1 (range: < 0.5-10) and 0.5 (range: < 0.5-7) IU/mL, were not significantly different from those of groups I and II. In conclusion, when antigenic value of rabies vaccine is less than 2.5 IU/dose (group III), the immune response is relatively poor. The doses of rabies post-exposure vaccination in children should not be reduced.