In two studies, we investigated the correlates of anthropomorphism, the attribution of unobservable mental states to inanimate entities and non-human animals. In Study 1, we investigated the relations between anthropomorphism, social understanding, empathy, prosocial attitudes, and history of childhood imaginary companions in a college sample (N = 264; M_age = 19 years, 2 months). In Study 2, we explored the relations between two different measures of anthropomorphism, theory of mind, imaginary companions, and social preferences in 73 children (M_age = 5 years, 5 months). Anthropomorphism was not strongly correlated with social understanding in adults or with theory of mind in children. There was, however, some evidence for links between anthropomorphism and reports of having imaginary companions and social preferences. Moreover, the two measures of anthropomorphism were not correlated with each other and yielded different patterns of results in Study 2, a finding that is discussed in relation to different forms of anthropomorphism. Statement of contribution What is already known on this subject? There are individual differences in anthropomorphism in both adults and children. There are two different methods to assess anthropomorphism: self-reports and narration of movies of geometric shapes. The two different methods that are used to assess anthropomorphism yield different developmental patterns. What does this study add? Individual differences in social understanding or theory of mind are not associated with individual differences in anthropomorphism. Having a childhood imaginary companion is linked with a tendency to anthropomorphize in both children and adults. Two measures of anthropomorphism are not correlated in children, consistent with the view that there are different forms of anthropomorphism.