Collecting intersection turning movement counts at intersections is an essential data collection task for many types of traffic engineering studies. On a daily basis, consulting firms and agencies collect turning movement counts that are used for preparing traffic impact studies, determining design hours for road improvements, and the retiming of traffic signals. Although there is a substantial body of literature on the stochastic nature of traffic volumes and how they vary by time, day of week, and season, the literature is quite sparse on how accurately turning movement counts can be collected at signalized intersections. The objective of this paper is to characterize the reliability of manual intersection turning movement counts performed with modern data collection technology.

Live intersection turning movement counts were performed on three days for three hours to characterize the range and reliability of percent errors. Every user, regardless of interface or device, improved from the first day to the second day, and all but one improved again between the second and third day. There was no clear superior data collection technology, but with the emerging ubiquitous of smart phones and tablets, the cost –benefit of these devices has the potential to change the manual counting techniques of the future. Traffic counter software running on smart phones or consumer electronic devices has the benefit of being affordable and perhaps more convenient as an electronic counting device (ECD).


Traffic Counting, Technology, traffic study, turning movements, data collection

Date of this Version