Impact of response enhancement methods on response rates, representativity, risk of bias and costs

Response rates play a key role in survey design and implementation and are the most widely-used indicators of survey quality. The pursuit of high response rates to minimise the threat of nonresponse bias continues to dominate decisions about resource allocation in survey research. Yet a growing body of research has begun to question this practice, highlighting the importance of all sources of survey error and the need to carefully examine how different types of error relate to survey costs. In this study, we use previously unavailable data from a new sampling frame based on Swiss population registers to assess the value of different methods of response enhancement designed to increase response rates on the European Social Survey. Using sampling data provides information about both respondents and nonrespondents, making it possible to examine how changes in response rate resulting from the use of different fieldwork methods relate to changes in the composition and representativeness of the responding sample. We compute an R-indicator to assess representativity with respect to the sampling register variables, and find little improvement in the sample composition as response rates increase. Closer examination of the coefficient of variation of subgroup response rates, however, suggests some improvements in the representation of certain subgroups as fieldwork progresses, and this is reflected in an overall reduction in the estimated Maximal Absolute Bias. We also estimated the costs of each fieldwork enhancement step, which gives a different perspective when evaluating the utility of these fieldwork enhancement methods. We discuss these findings in relation to the results of other similar studies exploring the potential and limitations of R-indicators, and attempt to draw conclusions about how to better target fieldwork efforts to reduce the risk of non-response biasstatus: publishe