The purpose of this study is to find out the effectiveness of self-support programs on the basis of participants’ income change and program satisfaction. For this purpose, we ran a t-test and a multiple regression analysis on the dependent variables of income change & participants’ satisfaction before and after participating self-support program. The participants' data set is from the "Study of Self-Support Program of low-income class", which is based on the panel data of 750 households from the first year to the third year. The dependent variables are the rate of income change of participants' and the degree of participants’ satisfaction. The independent variables are the types of self-support program. The variables such as gender, age, the level of education, the status of economic activity, the status of marriage, chronic disease, the period of participation in self-support programs are classified as control variable. A data set produced by Korea Institute for Health and Social Affairs is used to test the hypotheses. The result of this research is summarized as the followings. First, the labor market linking model compared with the labor support model significantly increased the income of the participants, while the model of human capital development significantly decreased the participants’ income. Second, each model compared with the model of labor support shows the following results in each area; model of occupation creation and the model of human capital development are more satisfactory in job hunting area; the model of occupation creation and the model of human capital development are more satisfactory in building participants’ job ability; the model of linking labor market and the model of occupation creation show lower satisfaction score in attaining psychological stability; the model of linking labor market results lower satisfaction in attaining self-esteem; the model of linking labor market and the model of occupation creation bring lower satisfaction score in helping make a living.