Abstract：The widening gap in societal and economic development between urban and rural communities and the relaxation of migration restrictions in China since the 1980s have led large numbers of rural laborers to leave their countryside homes in search of better job opportunities in urban areas. As a result of this wave of migration, millions of children were left behind by their migrant parent in rural communities, in the care of the nonmigrant parent, the children’s grandparents or other relatives, even some are left to care for themselves. Long-term and long-distance family separation and lack of face-to-face communications characterize the interactions between migrant parent(s) and their left-behind offspring, and this relative absence of parental affection puts left-behind children at a disadvantage compared with children from nonmigrant families. Although left-behind children were reported to have a higher probability of risk for maladjustment than their peers from nonmigrant families, literatures suggest that there are considerable variations in the adaptations of left-behind children; while some children are at a high risk for delays in adaptive functioning, some do not experience adaptive problems, and still some even show positive development. A key issue with important preventive and theoretical implications in the study of left-behind children involves the identification of potentially protective processes that support positive adaptation in children from migrant families. The present study was designed to examine the moderate effects of parental cohesion and children’s cultural beliefs about adversity on the relationship between peer rejection/acceptance and adaptive functioning by comparing left-behind children with children from nonmigrant families. A total of 424 rural children were recruited from a rural area in Henan province of China, including 76 children from two-parent-migrant families, 133 children from father-migrant families, and 215 children from nonmigrant families. Peer rejection and acceptance, children’s aggression and school disengagement were measured by peer rating. The participants also completed self-report measures of parental cohesion, Chinese cultural beliefs about adversity and loneliness. The results showed that peer rejection was positively associated with children’s aggression, loneliness and school disengagement, while peer acceptance was negatively associated with children’s loneliness and school disengagement. Father-child cohesion, mother-child cohesion and positive beliefs about adversity predicted lower levels of children’s loneliness. Moreover, father-child cohesion and mother-child cohesion respectively moderated the relationship between peer acceptance and children’s loneliness. Specifically, at lower levels of parental cohesion, peer acceptance was negatively associated with children’s loneliness, but such association was not significant at higher levels of parental cohesion. Additionally, the moderation effect of parental cohesion on the relationships between peer rejection and children’s aggression and between peer rejection and school disengagement varied according to the status of parental migration (two-parent migrant families and nonmigrant families). Compared with children from nonmigrant families, higher levels of father-child cohesion and mother-child cohesion better attenuated the relationship between peer rejection and the external problems among children from two-parent migrant families. These findings did not support the hypothesis that migrant parent serves as the by-stander in left-behind children’s development, and highlight the importance of parental cohesion and peer acceptance for the positive adaptations of left-behind children. The implications of these findings for interventions directed at left-behind children were also discussed.