EVALUATION WITH SPARSE NOMINAL DATA: THE CASE OF DIFFERENTIAL COMPLIANCE WITH THE 55 MPH LIMIT

This paper notes the inability of analysts to evaluate the impact of certain policy intervention because of sparse nominal level data. Using the premise that knowing the limits of what could have happened will assist the investigator to know what did happen, this research enumerated all possible solutions or, in real world terms, scenarios that represent changes in driving behaviour before and after the imposition of the 55 mph limit in 1974 on Michigan freeways. That enumeration tallied extreme values of all cells and some cell groupings in the 3 x 3 table representing the speed changes from 1973 to 1974. Enumeration of all possible solutions also yielded information about the likelihood that certain cells exceeded others. In some cases non-obvious and absolutely certain knowledge about interrelationships among some cells were uncovered. Substantively this investigation demonstrated that those 1973 drivers who travelled relatively fast modified their driving behaviour more radically than those who were closer to the new 1974 speed limit. (Author/TRRL)

  • Availability:
  • Corporate Authors:

    Pergamon Press, Incorporated

    Headington Hill Hall
    Oxford OX30BW,    
  • Authors:
    • Wanat, J
  • Publication Date: 1983-3

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00377305
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • Report/Paper Numbers: HS-035 061
  • Files: ITRD, TRIS
  • Created Date: Oct 30 1983 12:00AM