CW21 - 2021-03-30
Asparagus - CI1-CW21
Participants
- Jonathan Frawley / RSE, ARC, Durham University / e: jonathan.frawley@durham.ac.uk t: @df3n5
- Gary Leeming / CDC University of Liverpool / gary.leeming@liverpool.ac.uk
- Jez Cope / Data Services Lead, The British Library, Yorkshire / j.cope@erambler.co.uk / he, him / w: eRambler tw: @jezcope ma: @petrichor@scholar.social gh&gl: jezcope
- Kirsty Pringle / SSI Project Manager / k.pringle@epcc.ed.ac.uk
Context / Research Domain
User centred design principles for research apps / citizen science
Problem
Social science training is essential for designing effective questionnaires, information leaflets and other outputs designed for public consumption. In healthcare there is a similar discipline in public and patient involvement. However, there are no equivalent tools or widespread training in the usability of apps and websites, so software designed to collect data from non-specialists does not always work as intended and sometimes introduces unexpected bias. This can be a major issue in studies relying on citizen science or self-reported healthcare measures, for example.
Solution
Three overlapping parts to the solution:
- Resources: Create a resource of common design principles that can be published. Could be a chapter in The Turing Way, or a Carpentries-style workshop (or both).
- Tools: Creation and signposting of automated & semi-automated tools to support user centered design and identify potential problems in research instruments & methodologies.
- People: Build a network of people willing to test each other’s apps, instruments etc.
Diagrams / Illustrations
For tools, we could develop a website to check how our app / website might be used by people with certain disabilities, e.g - a colour blindness simulator (screenshots from https://www.color-blindness.com/coblis-color-blindness-simulator/ 2021-03-31)
Licence
These materials (unless otherwise specified) are available under the Creative Commons Attribution 4.0 Licence. Please see the human-readable summary of the CC BY 4.0 and the full legal text for further information.
Free-form notes from discussion
- Use of health (and non-health) data in interesting ways, e.g. using bin data to identify frailty risks
- Catalog of algorithms used by e.g. traffic cameras to improve transparency and trust
- Sensor data, air quality, blank template for usability?
- Can’t make data available for health/imaging so can’t make work reproducible.
- Periodic testing of reproducibility of published software / projects
- Lots of beginner training for librarians and historians, e.g. library carpentry. But get to intermediate, and bringing into practice. Building peer support network. https://glamdatasci.network/
- Opensafely.org
- See also https://www.datashield.ac.uk/ (Rebecca Wilson works on this project)