Sonification of Election Data
The aim of this sonification was to provide a representation of election results from the 2005 Styrian provincial parliamentary election, as an easily accessible and topical example of geographically distributed social data.
Our interest focused on the attempt to display social data, first, in their geographical distribution and, second, at a spatial resolution greater than usual. We wanted to design a sonification that would display spatial differences and similarities in the election results among neighbouring communities. The technique we chose was a model-based sonification approach.
The mental model behind our approach to the election data was that of a journey through Styria. A journey can be defined as the transformation of a spatial distribution into a temporal distribution. A traveller starting at community X passes first through the communities neighbouring X, and the longer the journey, the greater is the distance from community X. Hence, in this sonification, the distances between communities are mapped onto the time axis.
The 542 communities are displayed in a two-dimensional window on a computer screen (see Fig. 1). For each community, the coordinates of the community’s administrative offices were determined and used as the geographical reference point for the respective community. The distances as well as the angles within the data thus corresponded to the real distances and angles between the communities’ administrative offices.
The sonification is interactive: Clicking the mouse anywhere within it initiates a circular wave that spreads in the two-dimensional space. The movement of this wave is shown in the window by an expanding red circle. When reaching a data point, this point begins to sound in a way that reflects its data properties. In our case, these data properties are the election results within each community. The researcher can select one particular party at a time to listen to. Then, the pitch of the tone represents the percentage the respective party achieved at the election among a given community’s population: the higher the tone, the higher the percentage. Furthermore, the researcher can select the direction in which to travel. The line begins at the point where the researcher has initiated the wave. This line is used to support the perception of an aspect of the geographical distribution.
In a standard stereo speaker setting (or when using headphones), the communities are mixed to virtual spatial positions between the channels (‘panned’) according to their angle from the main axis: those located straight in front are represented by sound events panned to the centre, those at ninety degress left of the line by sound events from the left speaker, those in between to appropriate intermediate positions, and similarly for the right side.
Thus, the values for a given party can be displayed in about eight to ten seconds. In relation to browsing through the tables, this is a remarkable time gain. Also, in comparison with a standard two-dimensional graph, the combination of sound and interactive graphical display allows for rapid geographical localisation of data points or areas of particular interest.
The examples follow the direction indicated in the screenshot. Example 1 represents the results of the social democratic party in the 2005 provincial parliamentary election. The results, and therefore: the tones, are higher in the center of the country (traditionally industrial areas) while they are not as high in the beginning and in the end of the sound example.
Example 2 displays the results of the Bündnis Zukunft Österreich (BZÖ), the new party of the right populist Jörg Haider. Its results are relatively weak - only one community in the west of Styria noticeably stands out, which fosters the impression that this sonification is a helpful tool for quick outlier detection.
Dayé, Christian & Alberto de Campo (2006): "Sounds sequential: sonification in the social sciences." Interdisciplinary Science Reviews 31/4. 349-364.