Purpose: Specific language impairment (SLI) is known to aggregate in families. Debate exists on whether the male sex presents an additional risk for SLI. This meta-analysis examined whether there is a sex ratio difference in the risk for impairment among family members of an SLI proband and whether this is mediated by assessment method (direct assessment via psychometric tests vs. indirect assessment via questionnaire/interview) or relative type (sibling vs. parent). Method: Twelve studies met inclusion criteria, including 11 parent and 9 sibling samples. Risk ratios, indicating relative risk for language difficulties for males versus females, were calculated as a function of assessment method and relative type. Results: Direct assessments identified a male predominance of language impairment, with a pooled risk ratio of 1.73 for siblings (95% confidence interval [CI]: 1.20-2.52) and 1.54 for parents (95% CI: 1.14-2.07). No sex differences were observed for studies using indirect testing methods, with mean risk ratios of 1.12 (0.85-1.48) and 1.17 (0.92-1.49) for sibling and parent samples, respectively. Conclusion: A predominance of affected males among family members is observed when using direct assessments only. This finding is interpreted with reference to the strengths and weaknesses of different assessment methodologies and what sex differences may indicate about the biological mechanisms underlying the SLI phenotype.