The aim of this project was to replicate and extend findings from two recent studies on parent-child relatedness in autism (Beurkens, Hobson, & Hobson, 2013; Hobson, Tarver, Beurkens, & Hobson, 2013, under review) by adapting an observational assessment and coding schemes of parent-child relatedness for the clinical context and examining their validity and reliability. The coding schemes focussed on three aspects of relatedness: joint attentional focus (Adamson, Bakeman, & Deckner, 2004), the capacity to co-regulate an interaction and the capacity to share emotional experiences. The participants were 40 children (20 with autism, 20 without autism) aged 6–14, and their parents. Parent-child dyads took part in the observational assessment and were coded on these schemes. Comparisons were made with standardised measures of autism severity (Autism Diagnostic Observation Schedule, ADOS: Lord, Rutter, DiLavore, & Risi, 2001; Social Responsiveness Scale, SRS: Constantino & Gruber, 2005), relationship quality (Parent Child Relationship Inventory, PCRI: Gerard, 1994) and quality of parent-child interaction (Dyadic Coding Scales, DCS: Humber & Moss, 2005). Inter-rater reliability was very good and, as predicted, codes both diverged from the measure of parent-child relationship and converged with a separate measure of parent-child interaction quality. A detailed profile review revealed nuanced areas of group and individual differences which may be specific to verbally-able school-age children. The results support the utility of the Relationship Development Assessment – Research Version for clinical practice.