The report of intimate violence is often taken on face value in research. With no gold standard existing, one partner?s report is often an accepted measure of violence in the relationship. This discrepancy in reporting between intimate partners has been thoroughly examined, yet researchers still debate the reliability of single partner reporting. The current study extends upon the Archer (1999) meta-analysis of intimate violence discrepancy reporting, and includes current studies, instruments and research methods in the analysis. This study examines the discrepancy effect among thirty-four samples from studies of heterosexual intimate violence, finding that the discrepancy in general was sizeable for men?s violence, and that different study conditions yield differing levels of discrepancy. This discrepancy level is a major issue when conducting quantitative research, especially typological studies, which utilise estimation and extrapolation of prevalence levels, and when violence data is used to evaluate recidivism. The current study identifies situations where the discrepancy is likely to be high and makes recommendations to improve the accuracy of the data being used to advise government policy and spending.