Sustainability of voluntary travel behaviour change initiatives - A 5-year Study

During the early part of the first decade of the 2000s, a number of localities in Australia introduced Voluntary Travel Behaviour Change (VTBC) initiatives, otherwise known as TravelSmart. These initiatives were all monitored in the shortterm and suggested that there were reductions in person kilometres of travel (PKT) on the order of 6 to 18 percent. Beginning in 2007, the Institute of Transport and Logistics Studies (ITLS) was asked to undertake a 5-year study to determine if the effects of TravelSmart were sustained in the longer term. This paper describes the study methodology, which was a rotating panel drawn from the Australian Capital Territory, Queensland, South Australia, and Victoria, with panel members asked to carry a portable GPS device with them wherever they went for a period of 15 days in the September-November time period each year from 2007 to 2012, providing a total of six waves of panel data. All members of sampled households over the age of 14 were provided with a GPS device and asked to carry it with them. The paper reports on the inevitable panel attrition and the process of make up for attrition. The panel covered roughly 120 households per year, with approximately 40 households that had not participated in TravelSmart (the control group) and 80 households that had participated. Make up for attrition maintained this approximate split between the treatment and control groups. Details of the sampling procedures are provided in the paper. The sample provided data on about 3,600 person days of travel in each wave or a total of about 20,000 person days of travel over the six waves of the study. Each year, estimates were made on a state-by-state basis and for the entire sample of the change in daily average PKT for each of the control and treatment groups. The paper reports on these year-by-year averages for each of the two groups and for each state and overall. It was found that, while there was some variation from year to year, in general, the treatment group continued to show lower PKT than the control group, suggesting that the changes were sustained over the study period. To our knowledge, this is the first time that a longer-term monitoring of the effects of a VTBC has been undertaken, and is certainly the first one to use GPS measurements of travel to do this. The conclusions of the study suggest that the noncoercive policy of VTBC is effective for at least 5 to 8 years after the intervention is undertaken. This study also shows that a small sample of households using GPS for a two-week period provides an adequate basis for monitoring and evaluating the long-term effects of such an intervention.