Executive Summary

Overview Report

Main Report

Bibliographic History

Conclusion

"The general consensus in the scientific visualization field is that a broad commonality exists among the visual needs of all numerically intensive sciences. ...we are keenly awaiting its applications to fields with a shorter history in numerical computing, such as econometrics and the social sciences. Will users from these fields find this environment appropriate to their needs?"

Geography: of all the social science disciplines, geography would appear to make the greatest use of visualization techniques and technologies. Geography is closely affiliated with science, through links with physical geography and the earth sciences. This link has facilitated the flow of computer technology across the discipline in terms of both hardware and expertise. Geography is also relatively unique amongst the social sciences with its almost exclusive use of spatial data and its origins in the visual science and art of cartography.

Planning: the role of computer aided design (CAD) as a visualization tool in planning has expanded the range of alternative planning proposals under consideration in the professional planning arena. Computer-supported modelling and visualization may eventually serve as a common and efficient language, facilitating communication about multifaceted environmental planning issues. Within academia the principle research on the Web in this area concerns Virtual Worlds, and their role in visualising urban forms within multi-user environments. Such an interface, coupled with the accessibility of the Web, has the possibility of opening up a new paradigm within urban design that may be particularly applicable to widening public consultation and participation in development projects.

Psychology: similar to geography, psychology also shares the distinction of an interdisciplinary overlap between the sciences and social sciences. Current visualization techniques include various types of graphics such as contour plots, surface plots, scatterplot matrices and dynamic spinning. More recently, the use VR technology has started to supplement conventional research methods, such as in shape recognition and manipulation experiments which has traditionally used `physical' objects, but now also uses 3D interactive computer graphics. In numerous cases, computer visualization techniques has improved upon previous photographic and computational techniques

History: although history is not strictly a social science, a number of interesting interdisciplinary examples of visualization using historical data have been surveyed. There are visualization tools on the web which allow a simulation of a system that enables users to 'drill down' from the population as a whole, through the collective experience of members in a selected place to the histories of individual members. Other examples include Lexis pencils, LifeLine Projects and 3D Time-Space Visualization.

Politics, Economics and Sociology: despite being central to the social sciences, politics, economics and sociology appear to use very little of the cutting edge visualization technology available. One of the few exceptions in politics is the investigation of the political power of the media, and particularly television, in constructing and influencing global events. A notable exception for economics is the visual representation of the structures of world trade between twenty-eight OECD countries between 1981 and 1992. An innovative use of computer graphics in sociology is the visualization of social networks. Examples include visualising vistors' paths at Duisburg Zoo in Germany, resulting in interesting insights into the behaviour of visitors.

Social Statistics: in terms of complex visualization techniques one of the leading uses in social statistics has been in exploratory data analysis. This is becoming increasingly more essential as the typical social science dataset becomes more complex. Exploratory data analysis can act as a means of filtering extremely complex quantitative relationships among data into relatively simple, manipulatable graphical displays. This has allowed users to interact with their databases in real time, dramatically increasing the amount of information they can extract. However, the ability to easily query any part of the data set frequently can result in `information overload', which can often have detrimental outcomes.

In short, our review has taught us many things about visualization in the social sciences and we hope that its results will also be of more general interest and will change more widely held views. Firstly, there is no central core to this research and it is thus very difficult to define key research groups and centres, hence the need for organisations such as ACOCG, and interdisciplinary meetings within the social sciences. Consequently, most of the research is conducted largely in ignorance of much other work which either has already been done or which is currently being undertaken. Secondly, visualization in the social sciences is dominated by subjects with the closest links to the natural sciences or/and with a tradition in graphical output. A pattern of diffusion from science to social science is clear. Thirdly the World Wide Web is quickly becoming the dominant form of research dissemination as paper journals fail to evolve, charging exorbitant prices even for the production of simple two dimensional colour illustrations. We can expect all these trends to continue as there is no single discipline likely to dominate visualization in the future and so provide a core set of methodologies.

Our original specification asked for key research groups to be identified, but as we have stated above these are very difficult to define, given the diffuse nature of work in visualization in the social sciences. If we were to identify specific groups these would include: project ARGUS from Leicester, the SigGraph ACM, the University College Research on visualization in planning in London, the MIT Urban Studies and Planning Department, Alan Maceachren's centre in Penn State, and the work on visualization sociological networks in Germany. But overall it would be unfair to many other groups not to list them and to list all groups would not produce a key list!

Visualization in the social sciences continues to grow as a research activity beyond the original spurt of activity following the 1987 report. However this growth is relatively uncoordinated. The activity does not fall easily within the remit of any particular discipline and the publication of results in the traditional form is very problematic. This review has attempted to illustrate the wealth of work currently being conducted into visualization in the social sciences. It has provided a reference to the historical background through the creation of a very large bibliography of past papers and a long list of current web sites providing examples of different techniques and access to many different methods. Without greater coordination the future of visualization in the social sciences is likely to be much like the past, but more diffuse and more ephemeral. This coordination is likely to arise only from direct funding for exemplar projects and centres from the research funding councils.