As China's tax system is not perfect, the accurate information of household income and expenditure is more difficult to get.Thus, the identification of poor students in colleges and universities has become difficult to quantitative research.In this paper, nine non-income variables, as follows, the main domestic source of income, registered residence, poverty-stricken counties, single parents, major disease, grandparents income, the population of the schooling, family indebtedness, major incident, are usedto estimate the family's income and expenditure, and to identify poor students.The results show that the model is effective and feasible.