Cartograms, hexograms and regular grids: Minimising misrepresentation in spatial data visualisations

Abstract

The visualisation of spatial information is a powerful tool for researchers in urban analytics seeking to convey their findings to the wider research community and the public in an accessible way. Yet, even the most well-intentioned cartographer may introduce misrepresentation by mapping irregularly shaped and sized areas. This paper explores the extent to which different methods of visualising area-based data can remedy (or exacerbate) this misrepresentation by presenting results from a crowdsourced survey. Data from the 2016 European Union referendum at Local Authority level in England are visualised using four alternative methods (balanced cartogram, hexogram, hexagonal grid, square grid) and compared to a traditional choropleth map, in terms of people’s understanding of the authors’ intended message, through a crowdsourced survey questionnaire. Results indicate that mapping out original boundaries can introduce misrepresentation, which can be mitigated by using balanced cartograms and hexograms to improve the accuracy of visualisations.