Assessing the utility of private information in transportation planning studies: A case study of trip generation analysis

The extent to which private information should be used in planning for transportation infrastructure has not been fully resolved. To contribute to this debate, this paper discusses the argument for using private information by presenting a transportation planning case study in which the sample size required to detect a significant difference in two types of residential trip generation rates was reduced. One rate is based on ground counts and the other is based on household surveys. The number of vehicle trips generated by suburban dwelling units over a 24-h period is a critical element in the transportation planning process.When no-private information was used, the rates derived using the two methods were not statistically different. However, the use of private information might lead to different rates. Private information, described herein as individual transportation decisions, is defined by four characteristics: individually identifying (not anonymous); linked to another data source (not aggregated); available to a limited audience (not unlimited); and available for only a specific, socially legitimate use.The potential discrepancy between rates using ground counts and household survey data provides an opportunity to investigate the utility of private information. Simulations with an existing data set were performed to determine the extent to which additional private information could have revealed a discrepancy. The study concluded that the required sample size decreases substantially depending on the amount of private information available. The use of limited private information reduces the required sample size by over 30%. With more private information, the required sample size is reduced by over 90%. Thus, the use of robust statistical tests that is infeasible with only public information is possible with the availability of private data.The significance of this study is the approach presented to evaluate the utility of private information by documenting the tangible benefits, such as reduced sample sizes and reduced data collection costs, that will accrue if it is used. The political cost of acquiring the information can then be addressed, as can extending the approach to other situations where the use of private information is being considered.