STATEWIDE DISAGGREGATE ATTITUDINAL MODELS FOR PRIMARY MODE CHOICE (ABRIDGEMENT)

This effort which developed models using attitudinal and situational (demogrphic and trip-related) data collected in a home interview survey, had 2 objectives; to build mode choice models having both predictive and explanatory value, and to assess the relative significance of situational and attitudinal variables in regional or statewide models. Disaggregate binary logit models for the choice of bus as primary mode were constructed for 6 area-purpose cases. The results of the study suggest that attitudinal variables are less significant than situational variables in predicting primary mode choice. Data analysis showed that automobile users express more satisfaction with the attributes of transit than did those who actually used transit. The study concluded that small disaggregate data sets appear to be adequate for the development of useful mode choice models at the statewide level. Further conclusions regarding the independent variables best explaining mode choice, and the levels of influence of variables on mode choice are also presented.