Establishing and Using a Before and After Panel Survey: Case Study of New South Wales

Abstract

This paper describes the use of a panel in a pilot TravelSmart project in New South Wales. The survey was conducted using two-day diaries, for which households were initially contacted and recruited by phone, and then sent diaries in the post. The diaries were returned by post with two postcard and one phone call reminder. During recruitment, households were asked if they would be willing to do the survey a second time about 9 months later, and were recruited only if they agreed to do so. The second wave of the panel was initiated about 9 months later, and only households that had responded to the first survey were contacted again. The survey was completed in early March 2005. Out of 1107 households that completed the before survey, 776 (70 percent) completed the after survey. A significant change was found in vehicle kilometres of travel for those households that received TravelSmart tools, but there were no significant differences found in numbers of trips, or proportions of trips by either mode or purpose. However, the panel approach that was used in this study proved of considerable value. It is demonstrated in this paper that, without the panel, the same sample size would have yielded much less information about possible changes, or alternatively that a much larger sample would have to have been used, with concomitantly higher survey costs.