Legends of the dashboard: an empirical evaluation of split and joint layout designs for geovisual analytics interfaces

Figures & data

Figure 1. The tested tool with a multiple-legend layout (a) and a single-legend layout (b). Legend elements are marked with green areas: map key, metrics buttons, metrics drop-down list & timeline, county search (please refer to online colour version of the figure).

Table 1. Task types and questions used in the study.

Task type (Knapp 1995)	Question code	Question	Triad framework (Peuquet 1994)
			What?	Where?	When?
Identify	T1: identify1	What was the compensation payment per claim in Bergen in 2013 due to the storm surge?	searched	defined	defined
	T2: identify2	How many claims were submitted in Narvik in 2010 due to water intrusion?	searched	defined	defined
Locate	T3: locate1	Which municipality noted in 1994 the highest total compensation payment (in NOK) due to floods?	defined	searched	defined
	T4: locate2	Verify if the municipality with the highest compensation payment due to water intrusion in 2018 per capita in the county of Troms, has access to the seashore.	defined	searched	defined
Compare	T5: compare1	What temporal trend (increase/decrease) can you recognize in terms of total compensation payment due to storms in Sunndal?	defined	defined	searched
	T6: compare2	In 2000, extensive investments were undertaken in Oslo to mitigate flood risk. Can you see the impact of these actions reflected in flood-related metrics in the following years?	searched	defined	defined
Associate	T7: associate1	Looking at the whole country of Norway, can you distinguish regions with municipalities that repeatedly received higher values of total compensation payment with regard to storms in the period 2010–2018?	defined	searched	defined
	T8: associate2	Indicate one common characteristic in terms of location of the municipalities with compensation payment per capita above 1 K NOK due to flood in 2007.	defined	searched	defined

Figure 2. The procedure of the empirical study.

Figure 3. Participants’ average performance and confidence by legend layout.

Figure 4. Participants’ average performance and confidence by task (error bars: ± 2SEM).

Figure 5. The AOIs distinguished in multiple- (left) and single-legend (right) layouts.

Figure 6. Differences in the eye movement metric for tasks 1–8 between multiple- and single-legend layouts (Error bars ±2SE; ***p < .001).

Figure 7. Mean visit duration and visit count for each given task (error bars ±2SE; *p < .05; **p < .01; ***p < .001).

Figure 8. Transitions between elements of the interface in the multiple-legend layout (a) and the single-legend layout (b) (for clarity, the average number of transitions fewer than two are not presented). Own elaboration using a Circos Table Viewer (Krzywinski et al. 2009).

Figure 9. The average number of transitions between and within AOIs Views and Legends (error bars ±2SE; *p < .05; **p < .01; ***p < .001).

Figure 10. The way of setting the attributes from redundant interface elements. The missing share of 100% indicates the participants who failed in selecting the needed attributes and/or decided to keep the default settings.

Figure 11. The use of redundant elements, buttons and drop-down lists, in the multiple-legend layout.

Figure 12. The views from which the searched information was acquired.

Figure 13. Participants’ layout preference after solving the given tasks.

Figure 14. SUS scores for multiple- and single-legend layouts.

Figure 15. Criteria of selection declared while indicating preferred layout. Make note that the percentage regarding the two layouts referred to a different number of participants (75% of participants indicated multiple- and 25% of participants indicated a single-legend layout). Each participant could select several criteria.

Figure 16. Criteria of layout evaluation reported by participants.

Knapp, L. 1995. “A Task Analysis Approach to the Visualization of Geographic Data.” In Cognitive Aspects of Human-Computer Interaction for Geographic Information Systems, 355–371. Dordrecht: Springer. doi:10.1007/978-94-011-0103-5_25.

 Google Scholar

Peuquet, D. J. 1994. “It’s About Time: A Conceptual Framework for the Representation of Temporal Dynamics in Geographic Information Systems.” Annals of the Association of American Geographers 84 (3): 441–461. doi:10.1111/j.1467-8306.1994.tb01869.x.

 Web of Science ®Google Scholar

Krzywinski, M., J. Schein, İ Birol, J. Connors, R. Gascoyne, D. Horsman, S. J. Jones, and M. A. Marra. 2009. “Circos: An Information Aesthetic for Comparative Genomics.” Genome Research 19 (9): 1639–1645. doi:10.1101/gr.092759.109.

 PubMed Web of Science ®Google Scholar

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.