This article investigates unintended mobile access to surveys in online, probability-based panels. We find that spontaneous tablet usage is drastically increasing in web surveys, while smartphone usage remains low. Further, we analyze the bias of respondent profiles using smartphones and tablets compared to those using computers, on the basis of several sociodemographic characteristics. Our results indicate not only that mobile web respondents differ from PC users but also that tablet users differ from smartphone users. While tablets are used for survey completion by working (young) adults, smartphones are used merely by the young. In addition, our results indicate that mobile web respondents are more progressive and describe themselves more often as pioneers or forerunners in adopting new technology, compared to PC respondents. We further discover that respondents’ preferences for devices to complete surveys are clearly in line with unintended mobile response. Finally, we present a similar analysis on intended mobile response in an experiment where smartphone users were requested to complete a mobile survey. Based on these findings, testing on tablets is strongly recommended in online surveys. If the goal is to reach young respondents, enabling surveys via smartphones should be considered.
This study investigates online respondents' willingness to participate in research when using alternative data collection modes and techniques. The study compares mail surveys; face-to-face surveys; telephone interviews; and data collection techniques using apps, wearables, and register data. We also investigate whether respondents find certain topics relatively attractive to have measured in an alternative way compared to surveys. The results are based on a probabilitybased sample of about 2,000 Dutch respondents. The findings help to decide when to use different data collection methods for the social sciences. The results also provide insights into which respondents require extra persuasion to participate when new data collection modes are used.
With the growing popularity of smartphones and tablet PCs (tablets) equipped with mobile browsers, the possibilities to administer surveys via mobile devices have expanded. To investigate the possible mode effect on answer behavior, results are compared between a mobile device–assisted web survey and a computer-assisted web survey. First, a premeasurement in the CentERpanel is conducted to analyze the user group of mobile devices. Second, the users are randomly allocated one of the three conditions: (1) conventional computer-assisted web survey, (2) hybrid version: a computer-assisted web survey with a layout similar to mobile web survey, and (3) mobile web survey. Special attention is given to the design of the mobile web questionnaire, taking small screen size, and typical functionalities for touchscreens into account. The findings suggest that survey completion on mobile devices need not lead to different results than on computers, but one should be prepared for a lower response rate and longer survey completion time. Further, the study offers considerations for researchers on survey satisfaction, location during survey completion, and preferred device to access Internet. With adaptations, surveys can be conducted on the newest mobile devices, although new challenges are emerging and further research is called for.
Objectives Little is known about the 12-month prevalence of potentially traumatic events (PTEs) and to what extent the type of PTE is a risk factor for post-event lack of social support. In addition, it is largely unknown if pre-event mental health problems and loneliness, and demographics are risk factors for a lack of support. Aim of the present prospective study is to fill these gaps in evidence-based knowledge. Methods A survey was conducted among a large random sample of the Dutch adult population (i.e. the longitudinal LISS panel) in March-April 2018, and linked with pre-event mental health and loneliness data from surveys conducted in 2016 (n = 5,879). We distinguished four forms of perceived social support: emotional and esteem support, and social recognition and general disapproval. Results Loss of a significant other and/or colleague (28%) was the most prevalent 12-month PTE. The 12-month prevalence of violence, accidents and/or, and theft-related events was 13%. Multivariate logistic regression analyses revealed no differences in lack of emotional and esteem support, or in lack of recognition across non-death PTEs and death-related PTEs. However, victims of threat and physical (sexual) violence more often faced disapproval than
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.