Children pose a problem. The extended period of childhood dependency and short inter-birth intervals mean that human mothers have to care for several dependent children simultaneously. It has long been argued that this is too much of an energetic burden for mothers to manage alone, and that they must enlist help from other relatives to share the costs of raising children. Which kin help is the subject of much debate. Here, we review the evidence for whether the presence of kin affects child survival rates, in order to infer whether mothers do receive help in raising offspring and who provides this help. These 45 studies come from a variety of (mostly) natural fertility populations, both historical and contemporary, across a wide geographical range. We find that in almost all studies, at least one relative (apart from the mother) does improve the survival rates of children, but that relatives differ in whether they are consistently beneficial to children or not. Maternal grandmothers tend to improve child survival rates, as do potential sibling helpers at the nest (though the latter observation is based on rather few studies). Paternal grandmothers show somewhat more variation in their effects on child survival. Fathers have surprisingly little effect on child survival, with only a third of studies showing any beneficial effects. Overall, this review suggests that while help from kin may be a universal feature of human childrearing, who helps is dependent on ecological conditions.