Noncontacts

Noncontacts are a disposition that is used in telephone, in-person, mail, and Internet surveys both as a temporary and a final disposition. Two primary types of non-contacts can occur in surveys. The first type occurs when a researcher makes contact with a household or other sampling unit, and no one is present to receive the contact. The second type of noncontact occurs when a researcher makes contact with a household or other sampling unit, but the selected respondent is unavailable to complete the questionnaire.

For example, the first type of noncontact occurs during in-person surveys when an interviewer visits a household unit and finds no one there (but does find clear evidence that the unit is occupied). Noncontacts also occur when contact is made with a household or other sampling unit but the selected respondent is not available to complete the questionnaire at the time of contact. For example, this type of noncontact occurs with in-person surveys when an interviewer visits a sampled address, determines that the address is a household (or other sampled unit), administers the introductory script and respondent selection procedures to someone at the address, and then learns that the selected respondent is not available to complete the interview. This type of noncontact is very similar for telephone surveys and occurs whenever an interviewer dials a case, reaches a household, administers the introductory script and respondent selection procedures for the survey, and learns that the designated respondent is not available at the time of the call. Because contact has been made with someone within the designated sampling unit, cases that result in this type of noncontact usually are considered eligible cases and thus are included when computing survey response rates.

Noncontacts may also occur in mail and Internet surveys, but the nature of these surveys makes it very difficult for researchers to know when this is happening and makes it almost impossible to differentiate between the two types of noncontacts. For example, in a mail survey, the questionnaire may be delivered to a household when the residents are away for the entire field period of the survey. Similarly, in an Internet survey the respondent may be away from email and the Internet for the entire field period, or the questionnaire may be sent to an email address that the respondent does not check during the field period of the survey. Only if the researcher receives information (such as, in the case of an Internet survey, an automated email reply noting that a respondent is away) specifying that the survey questionnaire was sent to and received by the named respondent is the survey researcher able to determine conclusively that a noncontact has taken place.

Because noncontacts usually are considered to be eligible cases or cases of unknown eligibility (depending on the type of noncontact), researchers continue to process these cases throughout the field period. In order to better manage survey sampling pools, many researchers assign different disposition codes to the two different types of noncontacts. These disposition codes allow researchers to manage the sample more precisely. For example, noncontacts in which no contact is made with anyone at the household or other sampling unit often are recontacted on a variety of days and times (or after a specified period of time in a mail or Internet survey) to increase the chances of making contact with someone at the household or other sampling unit. For cases in which contact is made with someone in a household or other sampling unit (but the selected respondent is not available), the researcher can work to identify a good time to recontact the selected respondent. Because these types of noncontacts are a temporary disposition, it is important that researchers learn as much as possible about when to try to contact the selected respondent and then use any information learned to optimize the timing of additional contact attempts and, in doing so, to maximize the chances of converting the noncontact disposition into a completed interview.

...