The Effect of Cost and Importance Factors on the Return Rate for Single and Multiple Mailings

Abstract

Employing an experimental design, the effectiveness of cost and importance on the return rate for mail questionnaires is analyzed. Importance factors, notably certified mail, are shown to be of greatest consequence. Kent L. Tedin is an Associate Professor and C. Richard Hofstetter is a Professor in the Department of Political Science, University of Houston. Public Opinion Quarteily Vol. 46 122-128 ? 1982 by The Tiustees of Columbia University Published by Elsevier North-Holland. Inc. 0033-362X/82/0046-122/$2.50 1 Importance are distinguished from factors in that the latter indicates the salience of the substance of the survey to the person taking it. Herberlein and Baumgartner (1978) show that return rates increase if the content of the questionnaire is likely to be of interest to the respondent. Of course, investigators are not usually in a position to alter the content of a study for the purpose of increasing the rate of return. This content downloaded from 207.46.13.109 on Sun, 24 Apr 2016 06:35:09 UTC All use subject to http://about.jstor.org/terms SINGLE AND MULTIPLE MAILINGS 123 percent, respectively. Other importance found to be effective include the use of first-class postage, certified mail, and various methods of personalizing the communication, such as addressing the respondent by name and hand signing the letter (Blumberg et al., 1974; Linsky, 1975; Carpenter, 1974-75). Costs involve the time and effort a respondent must devote to completing the questionnaire. As costs go down, return rates should go up. In practice, cost and importance sometimes interact yielding results counter to expectations, For example, return rates should be higher for short than for long questionnaires-it costs less to complete the former. But long questionnaires may be seen as being more important, which counteracts some cost factors. Thus, Herberlein and Baumgartner (1978) find no zero-order relationship between questionnaire length and return rates, but find under controlled circumstances that longer questionnaires have lower return rates. The most effective cost reduction factor seems to involve offering monetary incentives. The respondent is compensated for the costs involved in filling out the questionnaire. Armstrong (1975), in a literature review, suggests a straightforward monotonic relationship between monetary reward and return rates-the greater the reward the higher the return rate. Others, however (Linsky, 1965; Schewe and Cournoyer, 1976; Cox, 1976), have suggested that certain discrete amounts are optimal. While considerable research has been done on mail questionnaire return rates, much of it involves a single contact, two-cell experiment which varies just a single factor, for example, the type of postage or the color of the paper. It may well be that a stimulus which is effective on an initial contact is not effective when used as part of a second or third contact. One exception to the two-cell experiments is an ex post facto design by Heberlein and Baumgartner (1978), in which the unit of analysis is the reported percentage return in other studies. The effect of more than one stimulus is then analyzed using regression. However, a more desirable approach to determining the effect of multiple contacts and multiple stimuli is a true experimental design. In this paper we report on an experiment involving two importance (multiple contacts, certified mail vs. first-class postage) and a cost factor (a monetary incentive) to determine the least expensive combination which yields the highest rate of return.