Article Preview
TopBackground
The theoretical approach of this study is to analyze the suitability of smartphones for participation in online surveys from a usability perspective, because from this point of view systematic processes and corresponding methods can be applied. For this purpose, the authors use the term usability as a theoretical basis to connect already known metrics and results of methodological research with the research field of human-computer interaction. Usability is defined in ISO 9241-11 as “the extent to which a system, product or service can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use” (ISO, 2018). For online surveys, the concrete context can only be determined to a limited extent in a quantitative study, since it can be assumed that there is an undefined situation in which a user interacts with a technical device. Also with regard to the concrete user, we can only expect that it may be a more or less typical participant. However, the components of efficiency, satisfaction and effectiveness can be derived and applied for a sufficient usability analysis.
A fundamental problem with online surveys on smartphones are high dropout rates (Sarraf et al., 2014; Mavletova, 2013; Lambert & Miller, 2015). Mavletova and Couper (2015) see a lack of opportunity to skip individual questions as a general reason for high dropout rates in online surveys, as this puts a burden on the participants. Tourangeau et al. (2013) also agree with this assumption, adding that data quality may be lower if answering all questions is mandatory. However, so far there are no studies examining this condition in more detail. Many dropouts and missing values can indicate dissatisfaction and thus a lack of usability.
Processing time is another problem of online surveys on smartphones. Some studies show longer processing time on smartphones (Lugtig et al., 2016; Mavletova, 2013; Horwitz, 2014). Especially when tables are used (Couper & Peterson, 2017) – a common form of presentation on smartphones – or if respondents have to zoom frequently (Olmsted-Hawala et al., 2016), mobile processing times are significantly higher. A high investment of time can therefore make processing inefficient and, as already mentioned, influence the usability component of efficiency negatively.