English title: Fieldwork details in the European Social Survey 2002/2003
Author(s): Achim Koch - Michael Blohm -
Type: Book chapter
The objective of the survey interview is to collect information about the population under study in a uniform and reliable way (Weinberg 1983: 329ff). For every survey data collection is a critical part of the survey process. In contrast to other parts of the survey process, like questionnaire development e.g., data collection is not in the hands of the researcher. In order to implement a face-to-face survey, the researcher usually has to cooperate with a survey organisation and its interviewers. In face-to-face surveys, the whole process of data collection consists of decentralised operations of lots of interviewers. That makes it often difficult to gain insights in what is actually happening in the field. Therefore, fieldwork sometimes has been called to be a "black box". This assertion holds for a survey in a single country. It holds even more, if we turn to cross-national surveys, where several countries and survey organisations are involved. Here, quite often only aggregate information about issues like the number of interviews realized, the length of the fieldwork period, etc. are available. The present paper tries to go a step beyond this. It provides more detailed information on several aspects of fieldwork in the European Social Survey (ESS). The ESS is a new multi-country biennial survey, which follows very high methodological standards and aims to improve survey methods and documentation of the whole survey process. In the years 2002 and 2003, the first Round of ESS was fielded as a face-to-face survey in 22 European countries. The paper starts with a brief introduction of the ESS in section 2. In section 3 we provide some basic data on fieldwork in ESS Round 1 (number of achieved interviews, length of fieldwork period, number of interviewers). In section 4 we proceed with information on various more detailed aspects of the work of the interviewers. The main focus is on the actual interviewing task of the interviewers, in particular the temporal aspects of their work. The issues analysed include the average number of interviews realized per day; the day and time when interviews are made; the length of the start-up phase until interviewers do complete their first interview; and the length of the time period interviewers are actually engaged in realizing interviews. With this information available, we can detect characteristic differences and similarities in the work of the interviewers across countries. Do these reported fieldwork details matter? In section 5 we do some exemplary correlational analyses. First, we examine whether there is a relationship between the fieldwork details described and the length of the fieldwork period in the ESS countries. Second, we investigate, whether there is a correlation between the timing of the interviews and data quality, concrete: the proportion of people in paid work included in the sample. In the conclusions in section 6 the need to replicate these analyses with other data sets is discussed.
From page no: 21
To page no: 51
Anthology: Conducting Cross-National and Cross-Cultural Surveys