Effects of smartphone use and recall aids on network name generator questions

Abstract

The increasing use of smartphones around the world provides new opportunities for network data collection using smartphone surveys. We investigated experimentally whether the use of smartphones and of a recall aid affects the number of reported names in a network name generator question. In a German online access panel (N = 3891), respondents were randomly assigned to answer the survey on their PC or on their smartphone and were randomly assigned to receive an open-ended recall aid question before the name generator question or after. Results showed that respondents on PCs and smartphones reported the same number of network contacts. This suggests that smartphone surveys have no negative effect on the network sizes in ego-centered network studies. However, requiring people to answer on smartphones resulted in a selection bias due to non-compliance, which may have led to an overrepresentation of persons with larger network sizes. The recall aid question did not lead to more reported names, but it proved to be an indicator of respondents’ motivation and response quality. In sum, the study suggests that smartphones can effectively be used for network research in tech-savvy populations or when respondents can choose to complete the survey on another device.