Conferences

As Biemer & Lyberg demonstrated in their book ‘Introduction to survey quality’ (2003) survey research has not missed the era of total quality management. Survey research can be viewed as a production process where each step should meet quality standards. Product quality – data quality in survey research – is achieved through process quality. Process quality depends on systems and procedures that are in place in an organisation.

In survey research using face-to-face interviews as mode of data collection the role of the interviewers is of crucial importance. The paper will focus on two different but linked types of interviewer evaluation. The first type is set up during the period of data collection and focuses – in real time – on the main tasks of the interviewers: a) the administrative part of the job; b) contacting respondents and getting participation and c) data quality of registered answers. The second type is a post factum evaluation of the data quality once the fieldwork is completed (e.g. response, item-nonresponse, and interviewer effects on data). Special attention will be drawn on (1) the preconditions to implement the evaluation programme; (2) the roles of the different actors during data collection: interviewers, the organisation responsible for the data collection and the organisation responsible for the whole survey process and (3) cost implications. The different parts will be illustrated with recent data from the survey on “Socio-cultural changes in Flanders”.
The Research Centre of the Flemish Government has been conducting an annual face-to-face survey on “Socio-cultural changes in the Flemish region and in Brussels” since 1996 (N= about 1,500 Dutch-speaking respondents, age 18-85 years, stratified two-stage sample design, named individuals from National Register). The survey is a tool to follow-up systematically changes in attitudes, opinions, behaviours of the Flemish people. General attitudes and opinions as well as attitudes and opinions towards the policy of the Flemish government are surveyed. From 2002 on ISSP modules are attached as a drop-off questionnaire. The personal interview contains a fixed part of questions and a variable part with a rotation of attitude scales and themes and lasts about 1h15min. The fieldwork force consists of about 100 trained free-lance interviewers.

A commercial marketing bureau is in charge of the data collection (public tender). Since the survey of 2000 a programme to prepare and to monitor the fieldwork and to evaluate the quality of the data has been set up. The Research Centre of the Flemish Government became responsible for the training of the interviewers, for a close follow-up of the interviewers and for writing standard reports such as a methodological report and an evaluation of the output and process quality.