Electronic Surveys

Electronic surveys have revolutionized the way evaluators and researchers collect information. The growth of Internet service providers has made electronic mail communication and the World Wide Web broadly accessible. The ubiquity of desktop and portable computing and Internet connectivity has prompted many evaluators to turn to these communication forms for data collection.

Electronic surveys appear on a computer screen just as they would in hand. Distributed as Web forms on the Internet, a unique location gives all respondents access the survey. Web-based surveys are developed with design tools and can accommodate any range of item types, including dichotomous answer questions (yes-no, true-false), multiple-choice, ranking and scaled response options, and open-ended questions. When the survey is complete, the respondent hits a “submit” button and the raw data are loaded into an active server page for later export or to database software (SPSS, Excel, Access, etc.) for manipulation or analysis. When electronic surveys are distributed as electronic mail messages, the preformatted document is completed and returned by the respondent as an e-mail message, with or without an attachment. In terms of convenience for the respondent and data collection and analysis compatibility for the evaluator, Web-based surveys are preferable.

Currently, vendors can produce Web-based survey software that does not require evaluators to also be Webpage designers. These products range from free Web-based shareware to packages costing several thousand dollars or more (i.e., Perseus, Remark, Questionmark, WebSurveyor, Super Collect). Higher end products have premium features that allow for branching, multisided table questions, and tabular display. Where survey data are heavily used in evaluation and no local expertise is available to develop customized software, some evaluation groups have made useful investments in Web-based survey software.

Key advantages to the use of electronic surveys include ease of survey editing and analysis, faster transmission and response time, and higher response rates. With an identifiable survey sample or parameter, tracking and follow-up is easier and less expensive. Research has also shown that electronic survey respondents may answer more candidly and provide more detailed responses to open-ended questions than respondents in paper surveys or interviews. There is some concern that due to the open nature of most online networks, it is difficult to guarantee the confidentiality of electronic surveys. However, many information assurance and security features make electronic surveying more reliable than postal mail surveying.

Respondents are limited to those with computer and online network access. This is especially problematic when working with low socioeconomic status or international populations. Also, constructing the format of a computer questionnaire can be difficult, leaving some evaluators to wish they had computer programming experience. Electronic surveys also usually require more respondent instruction and orientation.

Some electronic surveys, depending on system operation (server versus client based) can be very difficult to navigate. For example, a respondent may not be able to log off and then back on with answers intact. Using the back button on the toolbar or changing a previous answer may wipe out all previous answers. Occasionally, the print capability of electronic surveys may not function. As it is with technology in general, it is true with electronic surveys that there is a greater likelihood of glitches than with oral or written forms of communication.

...