Statistics South Africa (Stats SA) conducts both economic and social surveys. Within the Population and Social Statistics surveys cluster the following surveys are conducted: the General Household Survey, Income and Expenditure Survey, Quarterly Labour Force Survey, Domestic Tourism Survey, as well as ad hoc surveys and projects. For this paper the sources of response errors in these different surveys share common focal points:

The length of the questions during the design phase

Response errors introduced by interviewers

Response errors introduced by respondents

The paper to be presented will focus on innovative approaches that the Survey Operations division within Statistics South Africa has begun implementing. The division is currently capturing setup appointments and very recently piloting the Violence Against Women questionnaire using Personal Digital Assistant (PDA) technology. Lessons learnt from these new collection approaches will pave the way forward for the manner in which collection takes place in the future.

Key references:

American Statistical Association. 1993. How to Conduct Pretesting. Survey Research Methods. American Statistical Association.

Biemer, P.P., R.M. Groves, L.E. Lyberg, N.A. Mathiowetz and S. Sudman, eds. 1991. Measurement Errors in Surveys. John Wiley and Sons, New York.

Payne, S.L. 1951. The Art of Asking Questions, Princeton University Press, Princeton, New Jersey.

Tourangeau, R., L.J. Rips and K. Rasinski. 2000. The Psychology of Survey Response, Cambridge University Press, Cambridge, U.K.

Keywords: Control in the design and implementation phase of the survey cycle can minimize response error.

Biography: Mr Dion Pietersen is the Field Operations Manager for Statistics South Africa's Quarterly Labour Force Survey. This presentation will focus on work done over several years focusing on improving data collection in household based surveys.