Census overreporting inevitably occurs, including replicate enumerations within the census target population and mistaken reports outside the target population. The linear overreporting rate indicator estimator is currently the mainstream method for estimating census erroneous enumerations in some countries. An obvious defect of this method is that auxiliary variables are not used, which affects the accuracy of sampling estimation. Aiming at this problem, the research goal of constructing the estimator of the ratio overreporting index is proposed. In order to achieve the goal, a research method combining stratified double sampling, literature interpretation, and field investigation is adopted. With aim to facilitate readers’ understanding of research contents, empirical analysis is carried out by collected micro-actual data of the sample small census areas. The results show that the core criterion for judging a person’s overreporting population is not belonging to the target population, rather than his or her address registration error. The index estimator of rate overreporting based on stratified double sampling is suitable for census overreporting rate estimation. The sampling estimation accuracy of ratio overreporting index estimator is higher than that of linear overreporting index estimator. The existence of mistaken reports outside the census target population should not be ignored, otherwise the overreporting population will be underestimated. The rereporting within the census target population should not be confused with the reporting outside the target population. The significance of the research is embodied in three aspects. One is to establish a brand-new, operable, census index indicator system and its estimation theory system of overreporting rate census, which reflects the characteristics of overreporting rate of census, and improve or perfect the current theory of overreporting rate of census. The other is to analyze the overreporting of census from different perspectives according to this system, which would reveal the deep-seated reasons of overreporting of census, providing a reference for improving the operation method for the next census. The last is that the research results are expected to be applied to the overestimation of China’s 2030 census, so as to achieve the goal of scientific research serving social needs.
