This study examines gender differences in rural laborers‘ employment patterns in the mountainous and upland areas of Sichuan, China. The analysis employs both representative survey data of 400 households and geographical data. Multinomial logistic regression models are used to analyze the influences of gender, among other factors, on employment decisions of rural males and females, where the factors considered include personal, household, community natural environmental characteristics. Dividing laborers into four categories, we find that the proportions of males who participating in on-farm, pluriactive, and off-farm work, and unemployment were 24.41%, 28.64%, 46.27%, and 0.68% respectively, while that of females who participating in on-farm, pluriactive, and offfarm work, and unemployment were 43.20%, 13.95%, 30.95%, and 11.90% respectively. As to individual characteristics, age(AGE) and physical condition(PHY) effects appeared to be more pronounced for men, while education(EDU) and marital status(MAR)effects appeared to be more pronounced for women. Regarding household and community characteristics, the effects of the presence in the household of children aged 6-(CHI), number of persons in the household(POP), and labor force(LAB), per capitaincome in village(INCV), and the development status of village enterprises(ENT) on women were greater than that on men. In addition, the presence in the household of elderly individuals aged 65+(OLD) and time to reach the nearest township(TIME) are shown to have opposite impacts on men and women. While the presence in the household of pupils(PUP) and per capital gross value of industrial output(GVIO) was found to be irresponsive to men and women taking pluriavtivity and off-farm job. With respect to natural environments characteristics, the effects on men were opposite of those on women. Unemployment of women was found to be particularly responsive to household characteristics. A multinomial regression approach is undertaken to analyze rural males‘ and females‘ decisions of the four employment patterns considered, an approach that reveals considerable heterogeneity that is concealed by the dichotomous approach employed in most previous studies. The study thus contributes to our understanding of rural employment patterns and gender difference in mountainous and upland areas. This study examines gender differences in rural laborers' employment patterns in the mountainous and upland areas of Sichuan, China. The analysis employs both representative survey data of 400 households and geographical data. Multinomial logistic regression models are used to analyze the influences of gender, among other factors, on employment decisions of rural males and females, where the factors considered include personal, household, community natural environmental characteristics. -farm work, and unemployment were 24.41%, 28.64%, 46.27%, and 0.68% respectively, while that of females who participated in on-farm, pluriactive, and offfarm work, and unemployment were 43.20%, 13.95%, 30.95% As to individual characteristics, age (AGE) and physical condition (PHY) effects were to be more pronounced for men, while education (EDU) an Regarding household and community characteristics, the effects of the presence of the presence in the household of children aged 6- (CHI), number of persons in the household (POP), and labor force (LAB), per capitaincome in village (INCV), and the development status of village enterprises (ENT) on women were greater than that on men. In addition, the presence in the household of an elderly individuals aged 65+ (OLD) and while the presence of the household of pupils (PUP) and per capital gross value of industrial output (GVIO) was found to be irresponsive to men and With respect to natural environments characteristics, the effects on men were opposite of those on women. A multinomial regression approach is carried women. Unemployment of women was found to be particularly responsive to household characteristics. to analyze rural males 'and females' decisions of the four employment patterns considered, an approach that reveals substantially heterogeneity that is concealed by the dichotomous approach employed in most previous studies. The study thereby contributes to our understanding of rural employment patterns and gender difference in mountainous and upland areas.