Research On The Application Of Capture And Recapture Model In The Population Quality Evaluation Of Chinese Agricultural Census

2019 is a crucial year for a decisive battle to build a well-off society in an all-round way.The government attaches great importance to the development of the "three rural issues".Document No.1 of the Central Committee regards "realizing the priority development of agriculture and rural areas" as the development theme,and the "two sessions" also proposed the need to increase poverty alleviation and gave priority to the development of agriculture and rural areas.Among them,"village rejuvenation","precision poverty alleviation" and "population urbanization" have become hot topics in the two sessions on rural areas.In order to achieve these goals,we must have an accurate understanding of Chinese rural population,so that on the basis of grasping the base of the rural population,we can achieve accurate assistance to villages and households without omissions;at the same time,it can reasonably guide population flow,promote the development of population urbanization and gradually realize the great goal of revitalizing the countryside and building a well-off society in an all-round way.For rural areas,agricultural censuses are an important way to obtain comprehensive and detailed rural demographic data.However,the current Chinese agricultural census only aggregates the census registered population,which probably leads to a certain degree of deviation from the true value,lacking a credibility of the total rural population.At the same time,the current data quality assessment work only focuses on the acceptance of data in each link,which lacks a scientific theory guiding the evaluation of data quality,making it impossible to scientifically evaluate the accuracy of the total rural population.The article first combs the research status of the classic capture-recapture model in the field of demography and the application practice in various countries,and then explains the principle of the capture-recapture model and a commonly used construction method of population estimator under the theoretical framework of the model——post-stratification method.Next,it introduces the corresponding variance estimation method using the post-stratification method to construct the total population estimatorthe——the jackknife method,and clarifies the meaning of net error so that people can view the net error more rationally and more appropriately to use net errors to assess the quality of agricultural census population data.Finally,the article uses the third agricultural census data in Tianjin to demonstrate the poststratification method,and gives a rural population estimation method,a net error estimation method and a variance estimation method based on the capture-recapture theoretical framework to evaluate the quality of the census.The simulation study found that the estimated total rural population in Tianjin was 5,815,230 with a net error of-1.9% and a standard deviation of 20,410,25,indicating that Tianjin’s third agricultural census is of high quality.The significance of the article is to inspire the quality evaluation method of China’s agricultural census population data and provide a kind of ideas and methods for government departments to learn from.