Notice: the WebSM website has not been updated since the beginning of 2018.

Web Survey Bibliography

Title The Impact of Survey Communications on Response Rates and Response Quality
Year 2013
Access date 23.08.2016
Full text PDF (312 KB)
This paper investigates how efforts to improve response rates affect response quality in web surveys by assessing a number of indicators of quality. Response rates for all modes of survey administration have been declining in recent years and web-based surveys tend to have lower response rates than other administration modes. We conducted a survey to evaluate the services and supports provided by the Substance Abuse and Mental Health Services Administration’s (SAMHSA) Disaster Technical Assistance Center (DTAC). This study extends previous research that examined how the design of survey communication emails and landing page interact to affect survey response. The previous study found highest survey completion rates among those sent a long versus short email and presented with the web survey link at the top of the landing page versus the bottom. This may suggest respondents perceive survey burden only within the web survey interface and not the invitation email, and therefore may tolerate more information in email than on a landing page. The current study tests this by creating a 2x2 factorial design, where sample members are either presented the survey information needed for informed consent on a landing page (receiving a short email) or in an email (receiving a long email and bypassing the landing page), and either have the survey link/button presented at the top or bottom of the email communication. Maximizing survey response rates decreases the risk of nonresponse bias, helping to ensure that the survey will provide valid results. We compare response and completion rates and investigate whether the differences translate to variation in response quality by assessing response times, item non-response, patterns of systematic responses to grid questions, and use of the not applicable option when available. In addition we compared survey estimates across the four conditions to assess the impact on study findings. 
Year of publication2013
Bibliographic typeConferences, workshops, tutorials, presentations

Web survey bibliography - 2013 (625)