NORTHERN CALIFORNIA / SOUTHERN OREGON RURAL INTELLIGENT TRANSPORTATION SYSTEMS (ITS) AREAWIDE TRAVEL AND SAFETY IMPROVEMENT PROJECT. TASK 2.4 TRAVELER NEEDS SURVEY SIGNIFICANCE TESTING: VOLUME II

The demographic data collected in the Northern California / Southern Oregon Traveler Needs Survey were used to define the categories in the significance testing. Our analyses included significance testing based upon t-statistics to determine whether certain demographic variables (i.e. males, residents, specified age groups) significantly influence the response to various opinion-based questions. The chi-squared analysis is used to determine if differences in responses across groups (e.g., residents vs. tourists, males vs. females, or between specified age groups) are statistically significant at a predetermined level of probability. For this analysis, a 95% confidence level (p 0.05) was selected for reporting purposes. Thus, statistically significant differences in responses meant there was only five chances in one hundred that the variation across categories was due to something other than actual differences (e.g., biased samples) in the groups being analyzed. Before performing the modeling exercises to determine variable significance, minor data transformations were necessary. Data that had multiple non-numeric choices (i.e., urban/rural/suburban or work/school/shopping) were transformed into singular indicator variables. Data that had multiple choices with a range of numerical values were transformed one of two ways: (1) average values for each range were determined and assigned so that a single variable resulted for all values or (2) an indicator variable was created for each range of values so that multiple variables resulted. For example, a single new variable, avinc, was created by determining the average values for each of the income ranges specified. Interpretation of this variable led to conclusions about how increasing or decreasing income levels likely affect the opinion-based response. Nothing definitive could be said about any particular income group using this variable. Alternatively, four indicator variables (povinc, lowinc, midinc and highinc) were created for each income range. This specification allowed conclusions to be drawn specifically about certain income groups. Model interpretation is as follows. Inclusion of the variable in the model indicated that the variable significantly affected the response to the particular opinion-based question. For this analysis, a 95% confidence interval was used which corresponds to a t-statistic value of 1.96. Considering the example on the following page, familiar travelers and those traveling in the area for reasons related to school were found to significantly affect "Lack of information from signs along the roadway". The variable "one" is included in each model to allow for estimation of the bo value in the ordered probit model. Interpreting these results further, the negative sign preceding the estimated coefficient indicates a propensity to respond toward the lower or negative end of the ranking.

• Supplemental Notes:
  • This paper is on the CD-ROM, "Rural COATS (California Oregon Advanced Transportation Systems.")
• Corporate Authors:

  Western Transportation Institute

  Montana State University, Bozeman
  P.O. Box 174250
  Bozeman, MT  United States  59717-4250

  California Department of Transportation

  Division of New Technology and Research, P.O. Box 942873
  Sacramento, CA  United States  94273-0001

  Oregon Department of Transportation

  Transportation Building, Capitol and Center Streets
  Salem, OR  United States  97310

  Federal Highway Administration

  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Authors:
  • Harry, S R
  • Carroll, R W
  • Mounce, J
• Publication Date: 1998-8

Language

• English

Media Info

• Features: Appendices; Tables;
• Pagination: 75 p.

Subject/Index Terms

• TRT Terms: Age; Chi square distributions; Demographics; Gender; Improvements; Income; Intelligent transportation systems; Probits; Residents; Rural areas; Safety; Significance (Statistics); Statistical analysis; Tourists; Travel; Travel surveys
• Geographic Terms: California; Oregon
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management; Planning and Forecasting; Safety and Human Factors; Society; I72: Traffic and Transport Planning;

Filing Info

• Accession Number: 00942344
• Record Type: Publication
• Files: TRIS, ATRI, USDOT, STATEDOT
• Created Date: May 30 2003 12:00AM