Data Visualization

Chart Types and Methods

During the past decades, political science has accumulated a large corpus of various kinds of data such as comprehensive fact books and atlases, characterizing all or most of existing states by multiple and objectively assessed numerical indicators within certain time periods (e.g., OECD Factbook and Political Atlas of the Modern World). As a consequence, there exists a tendency for political science to gradually become a more quantitative scientific field and to use quantitative information in analysis and reasoning. Any analysis in political science must be multidimensional and combine various sources of information; however, human capabilities for perception of large amounts of numerical information are limited. Hence, methods and approaches for the visualization of quantitative and qualitative data (especially multivariate data) are an extremely important topic in political science. Data visualization approaches can be classified into several groups, starting from creating informative charts and diagrams (statistical graphics and info-graphics) and ending with advanced statistical methods for visualizing multidimensional tables containing both quantitative and qualitative information. Data visualization in political science takes advantage of recent developments in computer science and computer graphics, statistical methods, methods of information visualization, visual design, and psychology. Data visualization in political science has certain special features such as the frequent use of geographical maps, which creates a link with the well-developed field of geographic information systems (GIS). Furthermore, numerical tables in political science are often incomplete, which makes important the use of methods dealing with missing or uncertainly measured data entries.

There are two main types of numerical tables that can be the subject of data visualization. The first one is called an object–feature table, where each row represents an observation or an object and each column corresponds to a numerical feature or indicator commonly measured for the whole set of objects. An example of such an object–feature table [Page 538]is a fact book for a set of countries (see Figure 1), where the objects are countries and the features are numerical indicators such as gross domestic product (GDP) per capita, employment rate, life expectancy, and so on. The second type of numerical tables is called connection or distance tables, where both rows and columns correspond to objects and a numerical value characterizing a link between two [Page 539]objects is found at the intersection of a row and a column. Alternatively, such tables can be represented in the form of a list of links. A typical example of a connection table is the table representing migration rates or the mutual volumes of exports and imports for a set of countries (at the intersection of Row A and Column B, the volume of export from Country A to County B is found).

...