Concerns regarding response rates in survey research and the ethical use of financial incentives are ubiquitous. Pressure to obtain adequate response rates place nurse scientists in difficult dilemmas since low response rates are viewed as biased even when non-response bias analysis indicates the sample is representative (Asch, Jedrziewski, & Christakis, 1997). To assume that a study with a 60% minimal response rate is acceptable may be misleading since such studies, although perhaps less likely to have response bias, may still suffer methodological flaws (Lohr, 1999). Moreover, authors have argued there is no necessary correlation between low response and bias (Asch, Jedrziewski, & Christakis, 1997;Halpern & Asch, 2003;Halpern, Ubel, Berlin, & Asch, 2002).

With the demand for research participation high (Dresser, 2001), the difficulty of recruiting and retaining subjects is a critical concern. Successful nursing research depends on the willingness of subjects to participate in research, including, of course, willingness to respond to surveys and interviews. Most nurse investigators, however, have budgetary constraints and considerable financial compensation may not be a feasible option. But, nurse investigators continue to make important decisions about the type and amount of incentives that might be appropriate and effective in enhancing response rates. Data are needed to help guide these decisions and to address questions about the extent to which compensation should vary with research design, the sensitivity of the research questions, the specialty group being studied, and/or the degree of research risk.

Prima facie, providing financial incentives in nursing research is not morally problematic. Simply based on respect, it seems fair that participants receive some compensation in exchange for their time and associated burden. Further, it is sometimes acceptable or even appropriate to use cash specifically as an incentive. Many studies have reported significantly higher response rates using financial incentives without jeopardizing the overall quality of the study or individual item responses. On the other hand, excessive payment can raise concerns about undue inducement and subjects' voluntary participation. Just how much is enough, but not too much for responding to a survey-$1, $5, $10, $100? One model suggests paying subjects an amount similar to what they would receive for an unskilled job, but is this enough (Dickert & Grady, 1999)? Investigators are required to justify the use of incentives as well as the specific dollar amount to be used, yet there is variable guidance from Institutional Review Boards and limited empirical data to support specific cash amounts that might improve research response rates.

Current data support the fact that financial incentives improve response rates in research, although there is much to be learned about specifics. A great need exists for further ethical debate, empirical data, and critical review regarding use of incentives in nursing research.

References