Information Paper - Use of Incentives

INFORMATION PAPER

ISSUE: FVAP’s Use of Incentives in 2015 Survey of Registered Voters Living Overseas

BACKGROUND: The 2015 Survey of Registered Voters Living Overseas will be conducted in the spring and summer of 2015 to learn more about the demographics of registered voters living outside of the United States who requested an absentee ballot in the 2014 general election. This will allow for a more direct comparison between the voting behaviors of overseas citizens as compared to active-duty military and domestic civilian populations. Additionally, the survey will include questions about awareness and use of FVAP resources. It is anticipated that the results of this survey will allow FVAP to better understand the population it serves and to more effectively allocate resources.

A major threat to the accuracy of our survey data comes from the increasing prevalence of survey nonresponse, which raises concerns over the existence of nonresponse bias resulting from systematic differences between responders and nonresponders on key survey metrics. As a result, survey researchers have focused their efforts on finding effective methods of increasing response rates and minimizing the threat of nonresponse bias.

A large body of this research has focused on the use of incentives, both monetary and

nonmonetary, to encourage survey participation. Several studies have shown that a token

incentive (e.g., $2) sent with the questionnaire is often an effective means of boosting response

rates in mail surveys (Coughlin, et al., 2011; Church, 1993; Fox, Crask & Kim, 1988). In particular, Coughlin et al. examined response rates from a survey of recent U.S. veterans. They found that the response rate was 9% higher among those who received a $5 upfront incentive compared to those who did not receive an incentive.

Moreover, researchers suggest that the incentive provides a reason for participating in the survey among those who would not otherwise participate due to lack of interest in the topic of the survey (Groves, Singer & Corning, 2000). Therefore, including the incentive increases confidence that estimates are not biased by differences between responders and nonresponders. In our example, veterans who were satisfied with their service and have a positive association with the Military may be more likely to respond to the survey in the absence of an incentive, thus introducing bias into our estimates.

While considered a best practice in the private sector, the use of incentives in surveys conducted

by Federal agencies has raised a concern about the cost of implementing this methodology

relative to its effectiveness in improving survey quality. In response, the Office of Management

and Budget (OMB) released a January 20, 2006 MEMORANDUM FOR THE PRESIDENT’S

MANAGEMENT COUNCIL, Guidance on Agency Survey and Statistical Information

Collections, which recognized the effectiveness of monetary incentives for increasing response

rates (TAB 1). Moreover, Guideline 2.3.2 in OMB’s STANDARDS AND GUIDELINES FOR

STATISTICAL SURVEY released in September 2006 states that Federal agencies may use

incentives in order to achieve data of sufficient quality for their intended use (TAB 2).