Census can provide the fundamental population data of the whole nation. The census data are rich with hidden information that can be used for the investigation of national conditions and national power. Data Mining aims at extract the implicit, previously unknown, and potentially useful knowledge from voluminous, non-complete, fuzzy, stochastic data. Using Data Mining in census data can make full use of these data to provide services for country's social and economic development. Classification is one of the important Data Mining techniques. The Decision Trees of classification analysis can give high accuracy prediction results, and the output results are easy to understand. In this paper we use the Decision-Tree-Based classification model - CART to analyze the census data in Chengyang and Laixi, and classify the inhabitants, then evaluate the results, discuss the important significance for using Data Mining in census data.