Description
A study on four interaction and information presentation dashboard problems:
1. information overload,
2. inappropriate data order and grouping
3. ineffective data presentation
4. misalignment in visual literacy expectations
Each dashboard had two types: problematic and adapted. Then each type had two tasks.
File 1 (dependent and independent variables):
63 participants completed the experiment, so for each user, we have user graph literacy (0-4), effectiveness(0 or 1 for each task), efficiency (completion time in minutes for each task), perceived performance (0-20 for each task) and perceived cognitive demand (0-20 for each task).
File 2 (interaction data):
50 participants interacted with the dashboards while their low-level inttraction data was being collected using UCIVIT tool. The zip file contains 50 JSON files for each participant on eight dashboards.
File 3 (interaction patterns):
Using VMSP sequential pattern mining algorithm, we mined the low-level interaction data to find the patterns of behaviour exhibited by the participants. This file explains the benchmarking used to arrive at the parameters used and lists all the patterns found on each dashboard. It also further explains all the interpretations of events on all elements.
1. information overload,
2. inappropriate data order and grouping
3. ineffective data presentation
4. misalignment in visual literacy expectations
Each dashboard had two types: problematic and adapted. Then each type had two tasks.
File 1 (dependent and independent variables):
63 participants completed the experiment, so for each user, we have user graph literacy (0-4), effectiveness(0 or 1 for each task), efficiency (completion time in minutes for each task), perceived performance (0-20 for each task) and perceived cognitive demand (0-20 for each task).
File 2 (interaction data):
50 participants interacted with the dashboards while their low-level inttraction data was being collected using UCIVIT tool. The zip file contains 50 JSON files for each participant on eight dashboards.
File 3 (interaction patterns):
Using VMSP sequential pattern mining algorithm, we mined the low-level interaction data to find the patterns of behaviour exhibited by the participants. This file explains the benchmarking used to arrive at the parameters used and lists all the patterns found on each dashboard. It also further explains all the interpretations of events on all elements.
Date made available | 30 Mar 2023 |
---|---|
Publisher | University of Manchester Figshare |
Research Beacons, Institutes and Platforms
- Manchester Environmental Research Institute
Keywords
- interactive dashboards
- interaction dashboards
- information overload
- ineffective data presentation
- data ordering and grouping
- visual literacy
- interaction
- low-level interaction
- UCIVIT
- effectiveness
- efficiency
- perceived performance
- perceived cognitive demand
- NASA TLX
- NASA TLX task load index
- JSON
- interaction data
- human-computer interaction