Web Survey Bibliography
The Web has gained acceptance as a tool for survey researchers. It has also been found to have introduced a whole new gamut of potential cost and error trade-offs, and re-introduced long known trade-offs. As Web survey quality is explored and the Web is used to study new ideas and populations, there is at least one population where the Web survey has been found to be a very effective tool: college students. However, there remains significant work to fully understand how to most effectively use the Web when conducting survey research among college students. In our experiences with several multi-campus student surveys, the same implementation process and questionnaire has resulted in significantly different rates of response and survey completion from one campus to another. During the spring of 2004, the National Study of Living-Learning Programs (NSLLP) was conducted among undergraduate students at thirty-four college campuses. Response rates (AAPOR RR2) for each campus varied from 17% to 52%. Simultaneously, school administrators at each campus were surveyed about their institutions adoption of Internet technologies, student use of computing facilities, the general culture of email use at their campus, as well as other activities outside of the control of the research team that individual schools may have done to promote the NSLLP. In this presentation, we will describe what we found to be key characteristics about a college environment that results in high response rates to a Web survey. We will also describe what other institutional efforts were found to be effective at increasing response rates. We will discuss how our results may be used to tailor a Web survey data collection to fit specific campus characteristics.
Conference program