Evaluation of Electronic and Fax Methods for Survey Research

Abstract

In an effort to improve the efficiency and cost effectiveness of written surveys, we compared our experiences with distribution of surveys through e-mail and FAX to our earlier uses of surveys mailed in the traditional manner. We used a sample list of dietetic educators that provided e-mail addresses for about two-thirds of the members. If there was no functional e-mail address, we faxed the surveys. Respondents receiving the e-mail version were urged to return it electronically using their “reply” feature. If the respondents chose FAX or mail for their returns, they bore the expense. Using a system of two reminders and a modest incentive prize (awarded by lot), we achieved a 44% response rate within six weeks of the first distribution. With both e-mail and FAX it was possible to avoid making large numbers of copies of the four page survey, saving copying costs significantly. Each survey was individually addressed; this could be done without paper in the case of e-mail, but did require an individual cover letter for the faxed surveys. As was expected, since the electronic connectivity was already available, e-mail was the least expensive method. Skills and time required for handling differed between the three methods; faxing about 50 copies took several hours, time that could have been saved had we sacrificed the personalized introductory letter. We saw no evidence that respondents responding by FAX or mail were reluctant to bear the costs of their responses; in fact, some sent duplicate copies by more than one medium. The novelty of the delivery system may have been a factor; if e-mail or FAX surveys become commonplace, the responsiveness might diminish. The major shortcoming of e-mail is the limitation it places on sampling: lists including e-mail addresses are not readily available, many addresses include errors, use of listserves gives little control over the sampling, and one can not assume no bias between individuals using e-mail and those not. To a certain extent, the same sampling limitations apply to FAX surveys. Undoubtedly, as use of e-mail expands, high quality lists will become available; in that case, electronic surveys will probably be the method of choice for professional populations.