Fallback Statements

Oftentimes when interviewers first make contact with a sampled respondent, the respondent is hesitant or otherwise reluctant to agree to participate in the survey. In most surveys, researchers can anticipate the nature of the concerns that will be expressed by respondents, and, as such, the researchers can train their interviewers to use targeted persuasion to try to convince the respondents to cooperate. The verbiage that interviewers are encouraged to use to persuade respondents has been termed fallback statements, as the interviewers must "fall back" to them in order to be successful with the respondent.

For example, respondents sampled in RDD (random-digit dialing) surveys who express reluctance to participate often will ask the interviewer, "How did you get my number?" Knowing that this question is one that commonly is asked, the researchers can provide interviewers a suggested response that they can tailor to individual respondents. A fallback statement for this particular respondent question might be something along the following lines:

In addition to providing interviewers with a fallback statement to help them explain how someone's telephone number or address was chosen, other topics that interviewers commonly have to explain to respondents, and thus are ones that fallback statements should address, include (a) the purpose of the survey, (b) how the data will be used, (c) additional details about how the data will be kept confidential, (c) why only one designated respondent is chosen for each household contacted, and (d) who at the survey organization can be contacted if the respondent wants to verify the legitimacy of the survey.

If any of these concerns were expressed by the majority of respondents, then it would behoove researchers to build explicit details about it into the standard introductory spiel that interviewers provide to all respondents when they first make contact. However, not every respondent has such concerns, and even among those who do, they do not necessarily share the same concerns. Thus, the survey introduction needs to be a more generic one, with the interviewers having [Page 267]targeted fallback statements to deploy for those respondents who express specific concerns.