Home   About Kurtosis   Courses   Course Calendar   Booking Information   Ideas   Technique   Clients   Contact
Visualizing Statistics

Graphical explanations of statistical analysis

Visualizing Statistics is a one-day course that shows NHS information analysts how to draw charts that illustrate the results of statistical analysis. The course is all about the use of graphical techniques to communicate concepts visually. If you are looking for a course that concentrates on understanding the statistical method and calculations, you should try Demystifying Statistics.

Visualizing Statistics will suit information and performance analysts who want to learn how to use graphical techniques in Excel to communicate basic statistical concepts.

The first session of the course is called Visualizing Distributions and it deals with different ways of showing the full range of data rather than trying to summarize it by quoting the average.

We start by showing you how to create a perfect normal curve in Excel, then we take an anonymized NHS dataset and create visualizations based on the number of emergency hospital admissions per day. We look at different ways of drawing histograms (including taking a "population pyramid" approach) and boxplots. After combining histograms and percentiles together, we then we go onto looking at the number of bed days being generated by these admissions and draw cumulative frequency polygons to illustrate Glenday Sieve-type arguments.

The second session of the course is called Visualizing Time Series and it focuses first on ways of putting trends under the microscope by breaking them down into their quantiles.

We also look in this session at ways of adapting Kaplan-Meier survival curves to show trend data.

Visualizing Margins of Error is the third session. We explain how to draw caterpillar charts and funnel plots and discuss the key differences between them as ways of drawing attention to normal variation.

The final session of the course is called Visualizing Relationships. We cover scatterplots with single and multiple data series, scatterplot matrices and bubble plots.


To run Visualizing Statistics as an on-site course you will need either an IT training room with enough PCs for each participant or a meeting room big enough to accommodate the participants with a laptop each. We can bring laptops. Up to eight participants can be accommodated on each course.

To run Visualizing Statistics as an on-site course will cost 1,100+VAT. All expenses included.


"Yes, it's all very interesting, this, but I'm not sure I see the point of it..."

The course teaches the visualizations against the backdrop of the specific theme of emergency hospital admissions, so that the relevance of the techniques can be more easily grasped.

Every teaching example and exercise uses NHS data that has been used in real situations to shed light on real problems.

This is not a course about statistical graphics for their own sake; it is about using visualization to make sense of today's NHS issues.

Kurtosis 2012. All Rights Reserved.