Survey Response in RDD-Sampling SMS-Invitation Web-Push Study

Abstract

In this study, I tested a fairly new survey data collection approach, using random digit dialing (RDD) selection of mobile phone numbers combined with SMS invitations, with respondents asked to complete the survey online. The SMS included a URL that directed recipients to an online questionnaire consisting of primary and secondary socio-demographic questions, as well as items on the use of Internet, health, technology, and life satisfaction. The main aims of the study were to calculate and forecast response rate in a mobile web-push survey, and, through random assignment, to causally identify practices affecting response rates in survey research using this sampling type and the online survey mode. For this reason, a number of data collection characteristics were randomized and later used as predictors of survey (non)response. The results showed that response rates in a survey using the proposed approach are low, i.e., below 2% in these survey experiments. They remain lower than most other survey modes, probably below 5%, even when using an optimized survey design with pre-notifications, reminders, and the sending of texts to mobile numbers with appended geo-demographic information. These were the effective maximization methods and techniques identified in this study. The benefits of this approach to sample recruitment are its simplicity and cost-effectiveness, and its potential for use in the future by students, academics, and social and market research companies, as the nonresponse bias and coverage bias did not seem to exceed representation bias in similar probability-based surveys. Traditionally, cross-sectional general population surveys use many other recruitment approaches, for example, mail outs, telephone calls, or face-to-face contacts. Text messaging proved to be more than just an additional communication channel, or a medium for sending survey reminders.