Data visualization is the representation of data via graphical means. It uses symbols to represent quantities and categories. Good data visualization provides the context which allows us to quickly recognize patterns, intuit proportions, and make comparisons. In turn, these capabilities enable us to more easily spot relationships and trends in the data. Thus, visualizing data makes it easier for us to cognitively process the data and helps us uncover new insights.

ArcGIS Map Viewer

ArcGIS Map viewer provides charts which show distributions, relationships, and categories. Some of the chart types available include bar charts, histograms, and scatter plots which enable us to perform exploratory data analysis.

• A bar chart can be configured to display and compare multiple datasets either side by side, or as stacked bars. In addition, we can use 100% stacked bars to give a better sense of proportion.

• A histogram can be created for datasets with numeric values to show the distribution of the data.

• A scatter plot can show the relationship (correlation) between two variables. We can add a trendline to a scatter plot to highlight the relationship.

Data Clocks

Data clocks, which are available in ArcGIS Pro, are a type of data visualization that presents cyclical temporal data. Data clocks are especially useful for showing cyclical patterns. A data clock can help to uncover periodic patterns in temporal data, for example seasonal peaks and valleys. A data clock represents the data as a circular grid of cells in which individual records are binned into rings and wedges. Thus, data clocks aggregate data to help you understand general patterns over time.

Reachability Charts

A reachability chart is a type of chart that is automatically created by the density-based clustering geoprocessing tool. It shows the distance between points in the data set and their next closest neighbor. It plots the reachability distance on the y-axis against the reachability order on the x-axis. On the chart, valleys indicate clusters and peaks indicate noise points. One benefit of these charts is that they can help us understand how the method behind the tool works and fine-tune our analyses.

Conclusion

My first takeaway is that data visualization has a deep history that predates the use of computer graphics. This was illustrated by famous maps and charts made by John Snow, Florence Nightingale, and Charles Minard. More recently Hans Rosling has used computer graphics to present highly informative data visualizations.

My second take away is that effective data visualization works via pre-attentive cognitive processing. Good data visualization employs intuitive visual techniques that reduce the cognitive load on the viewer and enables them to perceive relative magnitudes, assess differences in magnitude, and spot patterns without having to hold information in memory and consciously process the data.