Non response in web surveys. The role of SMS reminders

Short Text Messages (SMS) are being increasingly used in different research contexts, including survey methodology. Although a number of studies investigated the effects of different strategies (e.g., the use of incentives or different contact modes) on response, there is not much research on the use of SMS reminders in Web surveys. Using experimental data from a national Web survey of Italian graduates in Social Work, this paper aims to evaluate the impact of SMS reminders on non response error at different stages of the data collection process. Drawing on the social exchange theory, we posit that SMS reminders may be more effective in boostering survey participation and data quality than E-mail reminders. Key findings are that: i) SMS have a positive impact on response rates, ii) although less marked, the differences in response rates between SMS and E-mail respondents persist at the end of the data collection process. We also discuss the specific challenges in evaluating the impact of different contact modes on measurement error e.g., the power of the analysis is limited by the low occurrence of inaccurate response behaviors that, together with the low response rates, reduce the sample size available for analysis.