Data from 520 British secondary school age children were used to explore determinants of and mental health outcomes (measured with the Strengths and Difficulties Questionnaire) from their non-resident fathers' relationships (child-reported father's involvement and frequency of contact) with them. Frequency of contact was negatively related to time elapsed since parents separated, and was higher if fathers were single than partnered. Compared to White, non-White adolescents reported lower frequency of father's contact but not lower father's involvement. Father's involvement was related positively to mother's involvement and negatively to interparental conflict. As a whole, non-resident father's involvement and frequency of contact did not improve the regression models predicting children's total difficulties or pro-social behaviour. Even after controlling for other factors non-resident father's involvement did not predict child's difficulties or pro-social behaviour, although frequency of contact was negatively associated with child's emotional symptoms. Neither child's ethnicity or gender nor mother's partner status moderated the link between child's mental health outcomes and non-resident father's involvement or frequency of contact.