Chapter 4 Geovisualization

In this first of three chapters dealing with mapping, I begin to explore the concept of geovisualization. Before getting into specific methods, I start with a brief discussion of the larger context of exploratory data analysis (EDA) and especially its spatial counterpart, exploratory spatial data analysis (ESDA). Central to this and to its implementation in GeoDa are the concepts of linking and brushing.

The technical discussion begins with an overview of key aspects of thematic map construction, followed by a review of traditional map classifications, i.e., quantile maps, equal interval maps and natural breaks maps. More statistically inspired maps are discussed in Chapter 5. Conditional maps are considered in the treatment of conditional plots in Chapter 8. The particular problem of mapping rates or proportions is covered in Chapter 6.

The common map types are followed by a discussion of various mapping options in GeoDa that allow interaction with the map window and the creation of output for use in other media. The chapter closes with an introduction of the implementation of custom classifications and the use of the project file.

Even though there is substantial mapping functionality in GeoDa, it is worth noting that it is not a cartographic software. The main objective is to use mapping as part of an overall framework that interacts with the data in the process of exploration, through so-called dynamic graphics (elaborated upon in Section 4.2). By design, maps in GeoDa do not have some standard cartographic features, such as a directional arrow, or a scale bar, since they are intended to be part of an interactive framework. However, any map can be saved as an image file for further manipulation in specialized graphics software.

As in the two previous chapters, the discussion here is aimed at novices, who are less familiar with basic cartographic principles. Others may just want to skim the material to see how the functionality is implemented in GeoDa. The treatment focuses on gaining familiarity with essential concepts, sufficient to be able to carry out the various operations in GeoDa. More technical details can be found in classic cartography texts, such as Brewer (2016), Kraak and Ormeling (2020), and Slocum et al. (2023), among others. Also highly recommended for those not familiar with mapping is Monmonier’s How to lie with maps (Monmonier 2018), which provides an easy to read overview of critical aspects of the visualization of spatial data.

To illustrate the various techniques, a new data set is used. It contains information on Zika infection and Microcephaly in municipios of the state of Ceará in northeastern Brazil. This constitutes a subset of the data for the whole of Brazil reported on in Amaral et al. (2019). In addition to the incidence of the two diseases in 2016, the data set also contains several socio-economic indicators from the Brazilian Index of Urban Structure (IBEU) for 2013 (see Amaral et al. 2019 for a detailed definition of each variable). Ceará Zika is included as a GeoDa sample data set.