The study used a cross-sectional retrospective design. The study used questions borrowed from the survey conducted by Peterson and Braiker (1980). In that survey, respondents were asked to indicate the total number of times they had committed each of 11 criminal acts during the 3-year period prior to incarceration. Official record data were obtained from inmate prison files. The majority of inmate prison files contained rap sheets and investigation reports. Generally, the analyses show that the validity and internal consistency of the self-report data for violent offenders are high and comparable with the findings obtained by previous researchers; however, the study confirms the finding that the worst offenders provide the poorest quality data. Implications for career criminal research and general measurement issues based on self-reported data are discussed. The findings suggest that the self-report measures of individuals lowest on self-control and highest on measures of antisocial behavior are likely to have poor validity and internal consistency. 6 tables and 40 references