A low-cost, prospective cohort study using the results of rapid diagnostic test performed at local clinics was conducted to estimate influenza vaccine effectiveness (VE) in school children (6-12 year-olds). All children in four primary schools in Tsuchiura City, Ibaraki, Japan were enrolled (n = 2607). Vaccination status and other risk factors were obtained with a baseline questionnaire. Participants were encouraged to visit a clinic to have a rapid test when they developed an influenza-like illness during the winter season in 2006-2007, and 88.6% of those who reported influenza to the school had been tested. The result of the test was obtained with another questionnaire. The attack rate of influenza A and B was 5.4% and 11.9%, respectively. Logistic regression was used to model the association between influenza vaccination and rapid-test-confirmed influenza after adjusting for potential confounders. Influenza VE was calculated as (1- adjusted odds ratio) × 100. VE for total influenza was 21% (95% confidence interval -8 to 42), which was a combination of VE for influenza A (44%, 8-66) and VE for influenza B (5%, -37 to 34). Among several possibilities that would account for rather low VE estimates in this study, low sensitivity of the rapid test, and differential propensity to seek vaccination or medical care between the vaccinated and nonvaccinated were considered to be important. This study was able to estimate influenza VE at very low cost with high specificity in case ascertainment by collecting the readily available data on influenza rapid test with questionnaires.