The aim of the current study was to examine the factorial structure of the Manchester Child Attachment Story Task (MCAST), using a father doll to address the child’s attachment representation to father. While the MCAST, a doll story completion task measuring attachment representations in early childhood, has been validated for use with a mother doll, its use for assessing attachment to father is relatively unexplored. Thus, an additional aim was to compare the factorial structure of the child’s attachment representation to father and mother, respectively. We analyzed data from 118 first‐grade children who underwent counterbalanced administration of the MCAST with a mother and father doll, respectively, within a period of three months. Exploratory factorial analysis revealed similar, three‐factor solutions for attachment to father and mother, with a first factor capturing the child’s (scripted) knowledge of secure base/safe haven and a second factor reflecting intrusive and conflict behavior. The third factor was different in the father and mother representations, capturing self‐care and role‐reversal in attachment to father and disorganization in attachment to mother. Findings support the potential usefulness of the MCAST for exploring the father‐child relationship and highlight a need for further research on early attachment representations to father.