Share

Social Survey Data Collection

2. Introduction

The challenge faced by all data collection agencies is to produce high quality and timely data that are collected in a cost effective manner.

In many respects this is becoming more difficult, not least because the long-term trend is of declining survey cooperation rates (Martin and Matheson, 1999; De Leeuw and De Heer, 2004).  All survey organizations work hard to counter this trend, whether through more sophisticated means of understanding response patterns and management of interviewer activity, better training of interviewers, or simply by paying respondents more to participate in their surveys.  In addition, there are some basic principles governing the organization of fieldwork that, if followed, will lead to improvements in the efficiency and quality of the data collected.  This chapter aims to outline these principles, based largely on observations made while working in the two biggest data collection agencies in the UK, the Office for National Statistics (ONS) and the National Centre for Social Research (NatCen).

Most of the observations relate to face-to-face data collection, principally with households or private individuals.  However, many of the recommendations can be applied in other settings, for example, data collection by telephone.  It has also been assumed that Computer-Assisted Interviewing (CAI) is the standard format for data collection.

In the broadest sense interviewers are there to:

...seek the cooperation of members of the public unknown to themselves to participate in voluntary surveys that can be about one or a number of subject areas on which they will not have any personal expertise.

Martin, J. and Matheson, J. (1999). Responses to Declining Response Rates on Government Surveys. Survey Methodology Bulletin, 45, July 1999.
De Leeuw, E. and De Heer, W. (2004). Trends in Household Survey Nonresponse: A Longitudinal and International Comparison. In: Groves, R. et al., (eds.) Survey Nonresponse. New York: Wiley.