EVALUATING DUI PROGRAM IMPACT IN QUASI EXPERIMENTS

This paper describes the limitations of quasi-experimental evaluation procedures frequently used in assessing driving under the influence (DUI) educational and remediation treatment programs. Although conventional designs for removing bias can be flawed, examples are cited in which relatively unequivocal inferences can be reached in terms of the presence or absence of a positive treatment effect. In general, it is believed that randomized experimental designs are superior to quasi-experiments and should be used much more extensively than has been the case.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00714620
  • Record Type: Publication
  • Files: TRIS, ATRI
  • Created Date: Dec 12 1995 12:00AM