SCV Design Development

Attachment 7. Development of the SCV Design
Traditional methods of interviewing include the in-person and over the telephone interviewing
strategies. There are strengths and weaknesses associated with both methods. Telephone
interviewing tends to be less costly than in-person interviewing as this mode does not involve
travel time to and from the household, and because telephone interviewers are generally paid less
than interviewers in the field. However, the telephone mode generally results in increased partial
interviews since respondents may easily break-off the conversation. Visual aids, which can be
helpful to respondents in remembering long lists, are difficult to incorporate in a telephone
interviews. Telephone interviewing is not conducive to administering lengthy informed consent
and assent procedures. Compared with face-to-face surveys, telephone surveys have also been
found to yield lower response rates (Groves and Kahn, 1979; Cannell et al., 1987; Sykes and
Collins, 1988; Hox and de Leeuw, 1994). Though in-person interviewing is more costly, this
form of administration generally results in higher response rates resulting from increased rapport
between the interviewer and respondents. Rapport in the face to face context also makes the inperson approach amenable to longer interviews. In-person interviewers can be cross-trained as
telephone interviewers, thereby eliminating the need for centralized, facility-based telephone
interviewing; however, traditional monitoring activities, one of the hallmarks of centralized
administration, are not feasible with this approach.
The self-administered modes, such as mail and Web surveys, tend to have lower per-unit costs
because no paid labor is involved—the costs for responding fall almost entirely on the
respondent (Groves et al., 2004). With their growing proliferation, Web surveys are increasingly
popular in mixed-mode surveys and are notably cost- and time-efficient (Dillman 2000; Couper
2000). The Web mode combines the advantage of computer-assisted response with the
advantages of self-administration, providing a data collection option that is both convenient for
respondents and cost-effective. With strictly cost in mind, a data collection approach that
emphasizes completion via the Web or mail is preferable. Research indicates that selfadministration elicits more honest reporting on sensitive topics than interviewer administration
(Tourangeau, Rips, and Rasinksi, 2000), but self-administered modes are generally characterized
by lower response and higher break-off rates compared to interviewer-administered modes (e.g.,
Gribble et al., 2000; Tourangeau, Steiger, and Wilson, 2002). Thus, they are often offered in
combination with other modes, such as CATI.
Given these mode considerations, and the objectives of the study, the SCV will deploy a mixedmode, multi-wave design that attempts to take advantage of the strengths of the modes while
recognizing their limitations. The design will blend a primary, interviewer-administered contact
mode for the household respondent (CAPI or CATI) with less costly options (inbound CATI and
Web) for (1) interviews with individual respondents in the household, (2) nonresponse follow-up
with household and individual respondents, and (3) interviews in subsequent waves. A mail
survey mode will not be utilized because of the challenges identified during the cognitive testing
of a paper-and-pencil version of the instrument.

Groves et al. (2004) identified three main reasons for using mixed-mode data collection: cost
reduction, response rate maximization, and money saving in longitudinal surveys. The use of a
combination of data collection methods reduces cost, as it typically involves an attempt to collect
data in a cheaper mode (e.g., Web), followed by a more expensive mode (e.g., telephone), and
possibly moving to an even more costly mode (e.g., face-to-face interviewing) for the
nonrespondent sample persons. Longitudinal surveys also employ mixed-mode data collection to
reduce cost in later waves, when rapport between the interviewer and the respondent has already
been established in the first wave, usually administered in face-to-face mode.
Moreover, one mode can be used to compensate for the weakness of another (e.g., Massey,
Marquis, and Tortora, 1982; Marquis and Blass, 1985; for a detailed discussion, see Groves and
Lepkowski, 1985). For example, in-person interviewing can overcome barriers to response
caused by not having a telephone number or households using call-screening devices to evade
interviewers. Mixed-mode designs are thought to promote response by providing respondents the
flexibility and convenience of choice, resulting in more opportunities to respond and in different
settings (i.e., at home, at work, or while travelling). By offering multiple modes simultaneously,
it is possible both to lower costs and to reduce nonsampling errors, such as nonresponse error
and measurement error (Biemer and Lyberg, 2003; Groves, 1989).