QUALITY AND COST-EFFECTIVENESS IN RECORDING A 24-HOUR TRAVEL DIARY

The current economic environment, together with an increasing sense of accountability, has meant that many authorities are searching for travel data which is not only cost-effective, but also retains the high quality required for planning and decision making. As part of a 1991 pre-pilot study for the Melbourne Area Personal Travel Survey, there was a detailed evaluation of different survey methods for collecting large- scale 24-hour travel data. One of the key objectives of this exercise was to design, implement, and recommend a methodology which gave high data quality, was cost efficient, and would be robust over an expected survey period of 5 years. Six different methods were tested. They included a version where a phone call was the initial contact, a series of methods where an introductory letter was the initial contact, and a personal interview approach. Other variations were the level of detail at which the travel data was collected (linked or unlinked trips) and the number of days over which the data was reported (one or two). The results were interesting at a number of levels. The telephone contact method gave neither better quality data nor was it any cheaper than the other methods. A consideration of the client's requirements for data usage indicated that the collection of linked-trip travel data would be inappropriate. Finally, where there were some relative cost savings in collecting two days' data from the one household, the question of how much extra data are actually obtained from the extra day remains a subject of debate. The paper provides detailed results in terms of actual and perceived quality of data and in terms of the cost of data collection, thereby making possible recommendations on cost- effectiveness of different survey methods. (A) For the covering entry of this conference, see IRRD abstract number 849214.