A Proposal for a Comprehensive Response-Rate Measure (CRRM) for Survey Research

Abstract

Respondents must co-operate with researchers. This truism is a fundamental proposition for almost all research into consumer and buyer behaviour where people are asked for information. Respondents or more correctly their response are the life-blood of social and marketing research whether that be conducted by the academic or the business community. However, it is a resource that appears to be showing some signs of depletion if not downright exhaustion. This paper conceptualises the activities of Interview Response Management (IRM) as a key role for project and field managers. The former trying to improve representativeness and the latter trying to manage the controllable interviewing factors that influence the level of response. Interest in the area arose from our investigation of trends in the levels of response and co-operation that the public were giving to market research surveys. It was immediately evident that reports of response rates in published survey data and even that collected directly from commercial market research organizations demonstrated a lack of consistency. There was no standard approach for defining response rate. Consequently we devised a new typology of response rate measures under the acronym CRRM – Comprehensive Response Rate Measure. This definition incorporated the key concept of respondent eligibility along with methods to estimate eligibility when the actual status of a particular survey contact was unknown. The approach focuses on the results of an initiated interview attempt and the variety of outcomes that may occur. In addition the CRRM was designed as an input to the process of IRM by providing performance indices that would be useful to fieldwork managers. It therefore examines 'surrogate refusal', call-back outcomes, nominal co-operation and survey participation rate. The derivation of this suite of measures is discussed in the paper along with the recommendation for its reporting in all published surveys. The authors contend that a standardised method of recording and calculating response rate will lead to a better understanding of the trends in survey co-operation as well as a more accurate comparison between initiatives for improving response.