To the editor: There has been a lot of research on incentives over the years; it is clear that they improve response rates to surveys, that nothing works as well as cash for this purpose, and that incentives often pay for themselves by reducing the number of attempts interviewers have to make to get people to respond. ("Your tax dollars at work: $5 bribes to take a government survey," Oct. 21)
One of the reasons all this research on incentives has been done is that fewer and fewer people are willing to take part in surveys. As a result, survey researchers around the world have turned to incentives as one of the most effective ways to stem this unfavorable tide. Nonresponse threatens to bias survey estimates.
The Consumer Expenditure Survey, the subject of the Gemini research, is one of the most important surveys done in the U.S. because the data feed into the calculation of the Consumer Price Index, on which many, many cost-of-living adjustments are based. So it is crucial that the survey is done right.
We should applaud the U.S. Bureau of Labor Statistics' efforts to maintain the survey at the highest possible level of quality, even if that means giving respondents some compensation for their time. It is hard to complete the Consumer Expenditure Survey questions; what's wrong with giving the respondents a token of gratitude for their time?
Roger Tourangeau, Rockville, Md.
Jessica Utts, Irvine
Tourangeau is president of the American Assn. for Public Opinion Research. Utts is president of the American Statistical Assn.