As part of the public involvement initiative undertaken for Pennsylvania's new long-range transportation plan (PennPlan), an attitudinal survey was administered to a sample of residents in the state. A few survey issues and an in-depth analysis of the survey questions are summarized. Information is provided on (a) the purpose of the survey, its components, and underlying themes; (b) the administration of the survey, its advantages and disadvantages, and possible solutions to improve the survey administration process; and (c) an in-depth analysis of the grades and importance ratings assigned to various aspects of the existing transportation system. The last issue receives particular emphasis. To compensate for the large number of nonresponses and the systematic response for the grades and importance ratings, the analysis was done in two steps. The first step uses sample selectivity to determine if systematic patterns existed between the residents who did and the residents who did not assign grades and importance ratings to certain aspects of the existing transportation system as functions of personal and residence characteristics. The results of this step were then used to "adjust" the coefficients in an ordered probit regression model that explains resident grades and importance ratings for the valid responses. By doing this, the models can account for the systematic nonresponse and yield models of responses on the basis of which unbiased inferences can be made.