Recent advances in multivariate methodology provide an opportunity to further the assessment of service offerings in public transportation for work commuting. We offer methodologies that are alternative to direct rating scale and have advantages in the quality and precision of measurement. The alternative of methodology for adaptive conjoint analysis for the measurement of the importance of attributes in service offering is implemented. Rasch scaling methodology is used for the measurement of satisfaction with these attributes. Advantages that these methodologies introduce for assessment of the respective constructs and use of the assessment are discussed.
In a first study, the conjoint derived weights were shown to have predictive capabilities in applications to respondent distributions of a fixed total budget to improve overall service offerings. Results with the Rasch model indicate that the attribute measures are reliable and can adequately constitute a composite measure of satisfaction. The Rasch items were also shown to provide a basis to discriminate between privately owned vehicles (POVs) and public transport commuters. Dissatisfaction with uncertainty in travel time and income level of respondents were the best predictors of POV commuting.
STEVEN SILVER, PhD
Steven Silver is a Professor in the Lucas Graduate School of Business and College of Business at San José State University. He has earned an MA and MBA from the University of Chicago, a PhD from the Haas School of Business, University of California, Berkeley, and has been a visiting scholar and post-doctoral fellow at the London School of Economics and at Stanford University. Dr. Silver has authored numerous reports and publications in consumer behavior, urban economics and measurement methodology. He has also served on advisory groups and panels for management of the arts and the design of transportation-related programs.
SJSU Research Foundation 210 N. 4th Street, 4th Floor, San Jose, CA 95112 Phone: 408-924-7560 Email: firstname.lastname@example.org