In many cases over the years, we failed to forecast accurately outbreak of Rice stripe virus (RSV) disease using population density of overwintering small brown planthopper (SBPH), Laodelphax striatellus. What is the source of error in this viral disease assessment? For answering the question, spatial distribution of RSV in paddy fields was investigated in relation to the yield of rice grain, and the population density of overwintering SBPH in ridge of paddy fields. Total 14 paddy fields in 5 regions were surveyed in June, 2008. Disease assessment was carried out with each of 30 rice plants in the one of the borders, 5th, 10th line from ridge, and 90 rice plants in diagonal line for conventional method. The ears of rices from 18 plants in the same surveyed line were collected in the late of August for the weights of 100 rice grains. The infection rate of RSV was decreased by the distance from ridge; on the contrary, the increase tendency was shown in the yield of rice grains in the middle of paddy field. It is suggested that the border effect of viral distribution between ridge and paddy field is caused by the spatial distribution of the vector insect. Thus, in this case, the narrow border effect could represent that the SBPH migrated short distance from the ridge to paddy field. The conventional method, observing the disease in middle area of diagonal line, showed no correlation to the population density of overwintering vector generation in ridge. It might cause underestimation of the regional disease rate. The counting of RSV-diseased plant in border line only overestimated it also. However, the RSV infection rate in the near border had highly positive linear relationship (R2=0.91) to the regional average values by regression analysis. Therefore, the near border sampling method can improve the RSV disease risk assessment.