Purpose: Social communication or pragmatic skills are continuously distributed in the general population. Impairment in these skills is associated with two clinical disorders, autism spectrum disorder (ASD) and social (pragmatic) communication disorder. Such impairment can impact a child's peer acceptance, school performance, and current and later mental health. Valid, reliable, examiner-rated observational measures of social communication from a semistructured language sample are needed to detect social communication impairment. We evaluated the psychometrics of an examiner-rated measure of social (pragmatic) communication, the Pragmatic Rating Scale--School Age (PRS-SA). Method: The analytic sample consisted of 130 children, ages 7-12 years, from five mutually exclusive groups--ASD (n = 25), language concern (LC; n = 5), ASD + LC (n = 10), social communication impairment only (n = 22), and typically developing (TD; n = 68). All children received language and autism assessments. The PRS-SA was rated separately using video-recorded communication samples from the Autism Diagnostic Observation Schedule. Assessment data were employed to evaluate the psychometrics of the PRS-SA. Analysis of covariance models were used to assess whether the PRS-SA would detect differences in social communication functioning across the five groups. Results: The PRS-SA demonstrated strong internal reliability, concurrent validity, and interrater reliability. PRS-SA scores were significantly higher in all groups compared to the TD group and differed significantly in most pairwise comparisons; the ASD + LC group had the highest (more atypical) scores. Conclusions: The PRS-SA shows promise as a measure of social communication skills in school-age verbally fluent children with a range of social and language abilities. More research is needed with a larger sample, including a wider age range and geographical diversity, to replicate findings.