Uncategorised / Thursday, 26 July, 2018

LIDA Seminar, Maps of time: Seeing population data as structures, not slices

With any LIDA seminar there is lots to take away from the session that is thought provoking. If you attended the most recent seminar in the LIDA series, concerned with data visualization and, in particular, novel ways in which population data can be mapped as structures, you were in for an interesting afternoon of reflecting on perception bias and how the human tendency to infer patterns intuitively from what we see can be used to improve data visualizations.

The thrust of the first half of the seminar, “Maps as Statistics? A call to Adventure for Perception Research in (geo)visualization”, delivered by University of Leeds School of Geography’s Dr Roger Beecham, was an examination of graphical inference: how we make sense of data visually and the pitfalls we encounter when it comes to visualizing spatial data in particular.

Read more about Dr Beecham’s talk here.

The keynote speaker for the seminar, Dr Jonathan Minton of the University of Glasgow, spoke on the theme of data visualization to argue for visualizing population data as structures as opposed to slices. In common with Roger’s presentation on LineUp tests, Minton started from a point of recognising the human ability to detect gestalt, i.e. the ability to perceive the whole as the sum of its parts. This laid the foundation for his argument that population data when displayed as a structure, gives the viewer access to more of the data and, crucially, means that they are better able to draw useful inferences from the data.

Read more about Dr Minton’s talk here.

Both speakers closed to some interesting questions from the audience: in respect of Beecham’s presentation weighing up the benefits of hex cartograms which of course remove the issue of irregularity as a feature of the data; and, in respect of Minton’s, how distracting colouration may be in population data structures because it privileges certain parts of the data by virtue of a more dominant colour having been selected.

This was the last LIDA seminar of this academic year. We’ve covered the following varied list of topics in this year’s series:

and we look forward to recommencing in September 2018. Keep an eye out on this page for news of upcoming seminars.

Feature image (c) Roger Beecham, and is an example of a Line-up test using Choropleth maps.