Research Article
A Study on User Acceptance of Error Visualization Techniques
@INPROCEEDINGS{10.4108/ICST.MOBIQUITOUS2008.3889, author={Hendrik Lemelson and Thomas King and Wolfgang Effelsberg}, title={A Study on User Acceptance of Error Visualization Techniques}, proceedings={1st International ICST Workshop on Human Control of Ubiquitous Systems}, publisher={ACM}, proceedings_a={HUCUBIS}, year={2010}, month={5}, keywords={context-aware computing pervasive computing location based services positioning error information visualization user study}, doi={10.4108/ICST.MOBIQUITOUS2008.3889} }
- Hendrik Lemelson
Thomas King
Wolfgang Effelsberg
Year: 2010
A Study on User Acceptance of Error Visualization Techniques
HUCUBIS
ICST
DOI: 10.4108/ICST.MOBIQUITOUS2008.3889
Abstract
Location-based services in general require information about the position of certain objects. For instance, for a navigation service the position of the user needs to be known. This position is usually provided by a positioning system. However, it is typical for all positioning systems that they are not perfect. This means that the positions they produce inherit position errors. Nowadays, usually only the position estimate is shown to the user even though a quality measure for the position error is provided by most positioning systems. To increase the user’s trust in location-based services and the usefulness of these services, the user should be informed about the uncertainty of position estimates as well. Thus, in this paper we investigate different visualization methods for the position and the position error. We carried out a user study to obtain information about the usefulness of the different methods. For this, we developed a questionnaire that contains nine different position and position error visualization methods. Furthermore, the questionnaire covers four typical application scenarios to be able to investigate whether users prefer different visualization methods for different applications. The results indicate that users are indeed interested in the position error they have to face. Further, they prefer a simple in-map representation of the position and the position error. These results are constant over different applications.