Many researchers have examined the effect of providing a variety of non-monetary incentives to subjects. These include token gifts, such as small packages of coffee, ball-point pens, postage stamps, key rings, trading stamps, participation in a raffle or lottery, or a donation to a charity in the respondent's name. Generally, but not consistently, non-monetary incentives have resulted in an increased response. A meta-analysis of 38 studies that used some form of an incentive revealed that monetary and non-monetary incentives were effective only when enclosed with the survey. The promise of an incentive for a returned questionnaire was not effective in increasing response. The average increase in response rate for monetary and non-monetary incentives was 19.1 percent and 7.9 percent, respectively.

Most researchers have found that higher monetary incentives generally work better than smaller ones. One researcher proposed a diminishing return model, where increasing the amount of the incentive would have a decreasing effect on response rate. A meta-analysis of 15 studies showed that an incentive of 25¢ increased the response rate by an average of 16 percent, and $1 increased the response by 31 percent.

A concern over the use of incentives is that the survey results may be positively or negatively biased because respondents are given an incentive to participate. Several studies have indicated that the use of incentives reduces to some extent item non-response and "bad answers," such as "don't know" or "no answer." It was also noted in a study published by Public Opinion Quarterly that respondents who received incentives have lengthier answers to open-ended questions. Though it seems likely that offering an incentive would bring apathetic participants to the study, research has proven otherwise. The data quality with an incentive, therefore, can actually be considered higher than if the incentive was not offered, as respondents have put more thought into answering the survey questions. There is also evidence that providing incentives will increase respondents' willingness to participate in future studies because they complete the survey feeling positive about the overall experience.

There is also the question of demographic bias. Is a certain group of people more likely to respond to a particular incentive, thus biasing the results toward that demographic group? For example, is it reasonable to assume that offering a monetary incentive would cause greater response from a low-income demographic group? It also makes sense that offering a beauty product would generate biased results because more women would respond than men. A research study conducted at Penn State University indicates, however, that the kind of incentive offered does not affect the response rate differently for different demographic groups. When presented with an incentive, people generally feel obligated to return the favour regardless of the type of gift they received. There is no indication of a group-specific effect of incentives.

There are situations, however, in which providing incentives to respondents may not be worth the additional response generated by the incentives. In the case of surveys about human rights abuses, the subsequent data may be used in a court of law to try perpetrators accused of crimes against humanity. If so, then defence lawyers might argue that respondents had been "paid" to respond in such a way as to put their client in a negative light. When considering the use of incentives, one must carefully weigh the benefits against the potential drawbacks.

1. This definition is comprised, in part, of subsets of the following web sites: www.nbrii.com/Customer_Surveys/ and www.statpac.com/surveys/incentives.htm (accessed 27 December 2006).