Noncontingent Incentives

Noncontingent incentives are traditionally used in survey research as a way of increasing survey response rates. The concept of noncontigent versus contingent incentives is that a noncontigent incentive is given to the respondent regardless of whether the survey is [Page 517]completed, whereas a contingent incentive is given contingent on the respondent's cooperation in completing the survey. Typically, the noncontmgent incentive would be given at the time the respondent receives the request to complete the survey. This type of incentive is most commonly used with mailed surveys, although it can be used in any survey mode. The most common type of noncontingent incentive in survey research is a monetary incentive paid in the form of cash or as a cash alternative, such as a check. The recent introduction of cash cards and gift cards have made them another viable option for monetary incentive use in surveys. Many nonmonetary incentives have been used to enhance response rates in surveys. Some examples of nonmonetary incentives that can be given as a noncontingent incentive include sweepstakes entries, videos, gas cards, coupons, online credits, small household appliances, books, electronic devices, small gadgets, and knickknacks.

Don Dillman advises that the proper use of non-contingent monetary incentives is one of the most important strategies a researcher can use to improve survey response rates. Social exchange theory postulates that small (i.e. token) noncontingent incentives make the respondent feel socially obligated, that is, “They already gave me something, so now I should do the survey for them.”

The scholarly literature shows a clear consensus that the use of a small noncontingent monetary incentive will increase cooperation rates in surveys significantly and is more effective than contingent incentives of considerably greater value. When considering which type of incentive, if any, to use in a particular survey, the researcher should consider the type of survey instrument (mailed, phone, Internet, intercept), the relative importance of the response rate, the level of effort required to complete the survey, the probable motivation of respondents, and the possible need to differentially incent members of some hard-to-reach demographic subgroups. For simple and short mailed surveys, short phone interviews, and short Internet surveys, an incentive is not likely to be needed. As the length and complexity of the survey increase or respondent engagement (e.g. level of interest) decreases, the need to consider the use of a noncontingent incentive is likely to increase. Care should be taken to ensure that the incentive offered is appropriate for the respondents being surveyed and does not introduce bias into the behavior of the respondent. An example of an inappropriate incentive would be a free DVD rental offered for participation in a television viewing survey. The respondent behavior that was being measured would most likely be impacted and the results may be biased.

The amount of incentive offered to the respondent should not be out of proportion to the effort required to complete the survey. A respondent who is given “too high” a noncontingent incentive amount as the sole motivating factor in the decision to cooperate in the survey may not answer the survey as accurately as someone else who received a noncontingent incentive of more modest value. Researchers should be aware of this buying cooperation phenomenon, which may cause some respondents to provide answers they think the researcher wants from them rather than providing accurate answers. Conversely, some respondents may have become so accustomed to receiving a noncontingent incentive when sampled for a survey that they may dismiss any survey request that does not offer one.

...