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

Web Survey Bibliography

Title Adding a Web Mode to Phone Surveys: Effectiveness and Cost Implications
Source The American Association for Public Opinion Research () 70th Annual Conference, 2015The American Association for Public Opinion Research (AAPOR) 70th Annual Conference, 2015
Year 2015
Access date 03.07.2015

High survey response rates are important when evaluating public health programs, and so are minimizing the associated survey costs. The goal of this study was to assess effectiveness and cost implications of collecting follow-up surveys using a dual mode survey (web and phone) compared to a single mode (phone). Data for the study come from a 2014 evaluation of Oklahoma’s Tobacco Helpline. A random selection of participants (n=1,174) were sampled for follow-up and contacted seven months after registering for services. Participants were not randomized into the two study protocols; participants with email addresses received dual mode protocols (n=467) while participants without email addresses received single mode protocols (n=707). Participants registered for services between March 1, 2013 and October 31, 2013. Overall response rate was 47.4% (556 out of 1,174). Effectiveness was assessed by comparing the survey response rates of single mode to dual mode study participants. A multivariate logistic regression model controlled for differences between the two survey groups. Cost implications were calculated by estimating what the cost would have been to conduct two separate surveys: a single mode and a dual mode; the costs of each survey strategy were then compared at different sample sizes. Cost estimates were categorized as fixed (e.g. infrastructure, interviewer training), varying by number of sampled participants (e.g. pre-notification letters), and varying by number of completed surveys (e.g. incentives). After controlling for co-variates, the dual mode survey protocols resulted in a higher survey response rate than the single mode (OR = 1.48, p=.009). However, the overall cost of a study was higher for the dual mode strategy until n=1,100 completed surveys were collected; at which point the dual mode strategy had lower overall costs. The cost analysis provides a helpful framework for assessing cost implications of new survey protocols.

Year of publication2015
Bibliographic typeConferences, workshops, tutorials, presentations

Web survey bibliography - The American Association for Public Opinion Research (AAPOR) 70th Annual Conference, 2015 (35)