Academic journal article Cartography and Geographic Information Science

Testing the Usability of Interactive Maps in CommonGIS

Academic journal article Cartography and Geographic Information Science

Testing the Usability of Interactive Maps in CommonGIS

Article excerpt

Introduction

High degree of user interactivity is a general requirement for maps designed to support "spatial thinking," i.e., hypothesis generation, data analysis, and decision making (MacEachren 1994; MacEachren and Kraak 1997). Recently, a number of software packages enabling user interaction with map displays have been created. Examples include:

* A dynamic link between maps and other types of graphical displays through "brushing;" objects selected in one of the displays are simultaneously highlighted in all of them (MacDougall 1992; Symanzik et al. 1996; Dykes 1997);

* Tools for changing, in real time, parameters of conventional cartographic methods, such as classes in a choropleth map (Egbert and Slocum 1992);

* Various interactive devices for controlling map animation (e.g., Edsall and Peuquet 1997; Blok et al. 1999; Harrower et al. 2000; Slocum et al. 2000).

Interactive techniques and tools can support information exploration and knowledge construction only when users are able to utilize these instruments properly. This problem is addressed in the research agenda of the ICA Commission on Visualization and Virtual Environments: "... users may find [new interactive techniques] difficult to apply, not derive the full benefit from them, or simply not utilize them if we do not consider various cognitive and usability issues" (Slocum et al. 2001, p.61). In particular, the Commission points to the necessity of evaluating methods via usability engineering principles.

In this paper we report on the testing of interactive tools implemented in our geovisualization package CommonGIS. Our hope, however, was not only to assess particular tools of CommonGIS but also to derive more general conclusions concerning people's understanding and adoption of interactive geovisualization techniques. More precisely, we wanted to check to what degree users are able to:

* Understand the purposes of geovisualization tools and learn how to use them;

* Retain the acquired skills after some period of not using them; and

* Develop a liking for the tools rather than being afraid of them.

We assumed that users would be quite able to master various tools for map manipulation but, possibly, only after receiving some training. Our study mostly confirmed these expectations.

Experimental Studies of Geovisualization Tools

Literature on the use of modern geovisualization tools that include interactive and/or dynamic map displays is rather sparse. One of the earliest studies was Koussoulakou and Kraak's (1992) experiment on the use of animated maps in comparison to static maps for analyzing spatio-temporal data. Koussoulakou and Kraak detected no major difference with respect to the users' ability to retrieve correct information and perceive temporal trends. However, the response times for animated maps were significantly (over 50 percent) faster than those for static maps. Koussoulakou and Kraak envisioned that the advantages of animated maps over static ones would be even more apparent for data with higher temporal resolution than was used in the experiment.

To keep their experiment simple, Koussoulakou and Kraak precluded user interaction with the maps. In contrast, Harrower et al. (2000) assessed interactive devices for controlling map animation. Specifically, they tested their implementation of the so-called "temporal focusing" and "temporal brushing" techniques (first suggested by Monmonier [1990]) in comparison to more traditional VCR-style buttons. The temporal focusing device was used to select time moments or intervals on a linear time scale, while the temporal brushing tool allowed choosing periods within the 24-hour daily cycle, thereby allowing the user to filter out diurnal fluctuations and look for long-term trends.

Surprisingly, Harrower et al.'s experiment did not show any superiority of the focusing and brushing tools over traditional controls in supporting pattern and trend detection and promoting hypothesis generation. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.