ANALYSIS OF THE COURT REFERRAL AND RANDOM ASSIGNMENT PROCESS: INTERNAL VALIDITY OF THE RESEARCH DESIGNS 1980 ANNUAL REPORT: VOLUME III

This volume of the 1980 Annual Report presents the final analysis of the CDUI Project's random assignment process. The treatment assignment procedures were detailed and sources of bias and controls were discussed. To verify that the random assignment process had distributed clients among the treatment groups of each research design in an unbiased manner, the groups were compared on fourteen client profile variables. The Project's three research designs (one for first offense drunk drivers, and two for multiple offenders) were examined separately. Thus a total of 42 comparisons were made. Of the 42 comparisons, only one resulted in a statistically significant between group difference. The magnitude of this difference was small and the statistical significance was considered spurious. Overall, the results of this study provided no evidence of bias in the random assignment procedures that could confound the analysis of treatment effectiveness. Within each research design, the treatment groups were comparable in terms of initial client characteristics.

  • Corporate Authors:

    Sacramento County Health Department

    Comp Driving Under Influence Alcohol Offender Treat Dem Proj
    Sacramento, CA  United States  95814

    National Highway Traffic Safety Administration

    Office of Driver and Pedestrian Programs, 1200 New Jersey Avenue, SE
    Washington, DC  United States  20590
  • Authors:
    • Reis Jr, R E
  • Publication Date: 1981-8

Media Info

  • Features: Figures; References; Tables;
  • Pagination: 91 p.

Subject/Index Terms

Filing Info

  • Accession Number: 00396147
  • Record Type: Publication
  • Source Agency: National Highway Traffic Safety Administration
  • Report/Paper Numbers: CDUI-AN-3-80 Final Rpt., HS-806 553
  • Contract Numbers: DOT-HS-6-01414
  • Files: HSL, TRIS, USDOT
  • Created Date: Aug 31 1985 12:00AM