Self-reports overestimate adherence compared to more objective measures such as electronic monitoring. However, self-report is the most feasible method for clinical settings; therefore, it is important to identify the context in which this method can provide an accurate assessment. To address whether self-reports are more accurate when missed doses are reported, we conducted a secondary analysis of data from a methodological study comparing multiple measures of adherence (including self-report and electronic monitoring) to a two-week placebo regimen mimicking HAART among 30 HIV-positive patients not on HAART. Results indicated a mean adherence of 85% and 62%, as measured by self-report and electronic monitoring, respectively. Self-report and electronic monitoring were not significantly correlated in the measurement of proportion of prescribed doses taken among the sub-group of 17 patients who reported missed doses (r = 0.22), nor among those who reported no missed doses, or the group as a whole. Using electronic monitoring as the validity criterion, these findings indicate that self-reports overestimate adherence even among patients who report missed doses.