Note: this project can be picked by two students simultaneously for the summer term 2023.
In recent years, researchers in psychology and other fields have attempted to reproduce the findings of some well-known studies, only to find out that in many cases the effects claimed by the original study could not be observed when the experiment was repeated. This failure to verify supposedly well-established knowledge was dubbed the replication crisis (Kosara & Haroz, 2018). Besides casting doubt on the original studies and the claimed results, this crisis shows that research is never “completed”, but requires continuous effort to verify and update findings in the context of societal changes and continuously developing research methods and tools.
Even if we do not doubt their results, many publications on historical studies have been found lacking essential information needed to replicate the experiment, which often has to be deduced from brief and potentially incomplete descriptions or graphical figures by researchers trying to recreate a study. The goal of this research project is to re-implement some classical user studies from cartography and related fields using a modern implementation framework, in order to make available the implementation as open source code, and to attempt to replicate the findings of the original studies in a modernized context (changed demographics, possibly changed display medium etc.). Furthermore, in addition to a replication as true to the original as feasible, the original study designs can be extended by additional aspects and parameters in order to broaden the generated insight besides the narrow replication effort – such as comparing desktop and mobile viewing scenarios, differentiation by age or other demographic factors, usage contexts, stress levels etc.
Candidates are invited to make their own proposal for which studies to replicate. In the scope of a single thesis project, a set of at least two related studies should be implemented and conducted (depending on the extent and complexity of the replication effort). Some examples for studies to consider would be:
(in alphabetical order; all papers are available from the supervisor upon request)
(proposals by the student for potential candidate papers to replicate are very welcome!)
Required knowledge and skills
The studies should be implemented using stimsrv (Ledermann, 2021; Ledermann & Gartner, 2021), a JavaScript framework for conducting user studies developed at the cartography group at TU Wien. A basic familiarity with JavaScript and web technologies (HTML, CSS, SVG) and a willingness to learn is therefore required for this project. An introduction to the stimsrv framework will be provided to candidates at the start of the project.
Staff working in this domain: Florian Ledermann (TU Wien)
Links
[1] https://gip.itc.utwente.nl/default/../research/msctopics
[2] https://gip.itc.utwente.nl/default/taxonomy/term/1548
[3] https://gip.itc.utwente.nl/default/taxonomy/term/1549
[4] https://gip.itc.utwente.nl/default/taxonomy/term/784
[5] https://gip.itc.utwente.nl/default/taxonomy/term/1550
[6] https://gip.itc.utwente.nl/default/taxonomy/term/498
[7] https://doi.org/10.1111/j.1467-8306.1971.tb00821.x
[8] https://doi.org/10.31219/osf.io/f8qey
[9] https://github.com/floledermann/stimsrv
[10] https://doi.org/10.5194/agile-giss-2-33-2021
[11] https://gip.itc.utwente.nl/default/research
[12] https://gip.itc.utwente.nl/default/print/research/msctopics/node/1302
[13] https://gip.itc.utwente.nl/default/printmail/node/1302