1504.0 - Methodological News, Dec 2008  
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 12/12/2008   
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product

Framework for household survey costs and fitness for purpose research

The Population Statistics Operations (PSO) area uses trained interviewers (also referred to as agents) to collect survey information for the Household Survey Program. A Household Field Cost Model (hereafter referred to as the model) has been developed to predict payment to agent costs up to four years in advance of survey enumeration. Increasingly, however, Household Survey areas are also asking questions such as:

    • What if I conducted part of my survey as a telephone interview and part of it face-to-face?
    • How would my costs and quality of output change if I were to increase or decrease my target response rate?
    • How should I run my survey to achieve the results that I want for the least cost?
Questions such as these are not readily answerable from the model as they involve the interrelationships of a number of variables. Respondents are not all the same and neither are interviewers and these factors play a significant part in the tension between cost and survey outcomes such as response rates and contribution to estimates. The Operations Research Unit has commenced a project to develop a framework for managing the trade-off between field costs and data quality to answer questions such as those raised above.

The first phase of the project was to review the current model to determine its ability to accurately predict costs across the range of surveys in the Household Survey Program. This phase identified some quick modifications to the model to improve the accuracy of household survey cost predictions in the short term.

The next phase of the project, which is currently underway, is aimed at developing a mathematical model for costs which is transparent and accurately reflects costs of operational procedures involved in data collection. The model will be able to assess 'what if ' scenarios, specifically including different field procedures; and will be able to be readily maintained and used for monitoring costs. Work is also underway to develop measures of fitness for purpose for time series and cross sectional estimates, and to understand the relationship between operational procedures and fitness for purpose.

For more information on this project, please contact Louise Gates on (02) 6252 6540 or louise.gates@abs.gov.au.