Journal of Quantitative Criminology

, Volume 16, Issue 1, pp 87–110

Comparing Self-Reports and Official Records of Arrests


  • Michael G. Maxfield
    • School of Criminal JusticeRutgers University
  • Barbara Luntz Weiler
    • School of Criminal JusticeUniversity at Albany
  • Cathy Spatz Widom
    • School of Criminal JusticeUniversity at Albany

DOI: 10.1023/A:1007577512038

Cite this article as:
Maxfield, M.G., Weiler, B.L. & Widom, C.S. Journal of Quantitative Criminology (2000) 16: 87. doi:10.1023/A:1007577512038


We compared measures of self-reported arrests and official arrests for 676young adults with a history of child abuse and/or neglect and 520 nonabusedand nonneglected controls matched on age, sex, race, and approximate familysocial class. Findings reveal considerable concurrent validity between thetwo sources overall. But there is also evidence of differences by gender,race/ethnicity, age at time of arrest, conviction status, and type ofoffense. Abused and neglected subjects did not appear to differ from thecontrol group in the extent of underreporting of known offenses, however,the groups did differ in the degree of “positive bias”—offensesnot found in arrest records. Abused/neglected subjects self-reportedproportionately more offenses not known to police compared to controls. Thisresult suggests that findings from previous studies on the relationshipbetween childhood victimization and later criminality, as measured byarrests, may have underestimated the magnitude of this relationship.

self-reportsofficial reportsvaliditychild abusechild neglect

Copyright information

© Plenum Publishing Corporation 2000