Abstract: To study the sustainable livelihood is significant for poverty alleviation and the development of rural area. Based on the sustainable livelihood framework, 484 farmer householders were investigated in Shangluo City of Qinling-Daba mountainous area, Shaanxi Province, China. The survey data was used to analyze household livelihood capital between different groups by constructing the indexes system of household’s livelihood assets, then the impact of livelihood capital on the livelihood strategies as well as their coupling coordinative degree were discussed using the Multinomial Logit regression model and the livelihood capital coupling coordinative degree model respectively. The results showed as follows: (1) The livelihood strategies of rural households were obviously different. According to the proportion of non-agricultural income, there were four types of household livelihoods, namely, exclusively employed by others (Type A),employed by others most of the time (Type B),work part time for others (Type C) and 100% doing the farm work (Type D). (2) In the survey samples, the livelihood capital of farmers was limited and unbalanced, which presented the characteristics of relatively high financial capital and social capital and relatively low natural capital and human resources capital. Among the non-poor households, the total livelihood capital of the Type C was the highest (0.451), followed by the Type B (0.393) and the Type A (0.382), the Type D (0.215) was the lowest. While among the poor households, the Type B (0.348) was the highest, followed by the Type A (0.345) and the Type C (0.342),the Type D (0.184) was the lowest. (3) The impact of livelihood capital on the livelihood strategy choice of non-poor households and poor households was different. For the non-poor households, the per capita cultivated land, the per capita forest land area, the cultivated land quality, vocational skill level, political resources and the employment network had a positive effect on the transformation from the Type A to the Type B and the Type C, but per capita household income and male labor ratio had a negative effect. The per capita household income and vocational skill level had a negative effect on the transformation from the Type A to the Type D. For the poor households, the transformation from the Type A to the Type B, Type C and Type D was positively impacted by the per capita cultivated land, the per capita forest land area and political resources, and the per capita household income, labor education level, vocational skill level and communication expenditure were negative factors.(4) If lining the coupling coordinative degree of non-poor households' livelihood capital up in order, from the largest to the smallest we have the list as follows: the Type C (0.114),the Type B (0.106),the Type A (0.103),and the Type D (0.045).Similarly for the coupling coordinative degree of poor households’ livelihood capital we have the list as follows: the Type C (0.095),the Type A (0.094),the Type B (0.092) and the Type D (0.086).This study could provide useful information for the optimization of livelihood strategies and effective poverty alleviation.

Key words: household, livelihood capital, livelihood strategy, Qinling-Daba mountainous area