Blog.

Visualization.

How Correspondence Analysis Works (A Simple Explanation)
19 May 2017 | by Tim Bock

Correspondence analysis is a data science tool for summarizing tables. This post explains the basics of how it works. It focuses on how to understand the underlying logic without entering into an explanation of the actual maths.    …

Continue reading

correspondence analysis movements How to Interpret Correspondence Analysis Plots (It Probably Isn’t the Way You Think)
19 May 2017 | by Tim Bock

Correspondence analysis is a popular data science technique. It takes a large table, and turns it into a seemingly easy-to-read visualization. Unfortunately, it is not quite as easy to read as most people assume. In How…

Continue reading

Easily Add Logos to a Correspondence Analysis Map in Q
17 May 2017 | by Tim Bock

You can take your correspondence analysis plots to the next level by including images. Better still, you don’t need to paste in the images after the analysis is complete – you can include them right…

Continue reading

correspondence analysis Easily Add Images to a Correspondence Analysis Map in Displayr
17 May 2017 | by Tim Bock

You can take your correspondence analysis plots to the next level by including images. Better still, you don’t need to paste in the images after the analysis is complete – you can include them right…

Continue reading

correspondence analysis Easily Add Images to a Correspondence Analysis Plot in R
17 May 2017 | by Tim Bock

You can take your correspondence analysis plots to the next level by including images. Better still, you don’t need to paste in the images after the analysis is complete – you can include them right…

Continue reading

Cluster analysis 5 Ways to Deal with Missing Data in Cluster Analysis
10 May 2017 | by Tim Bock

If you have ever tried to perform cluster analysis when you have missing data, there is a good chance your experience was ugly. Most cluster analysis algorithms ignore all of the data for cases with any missing…

Continue reading

pictograph_vs_pie_vs_bar Where Pictographs Beat Bar Charts: Proportional Data
03 May 2017 | by Carmen Chan

Pictographs are exceptionally good for some types of data. In my earlier post, I discussed how they are great for showing counts. In this post, I show how useful they are for displaying proportions (e.g. rates, percentages, fractions)….

Continue reading

Pictograph of alcohol consumption Where Pictographs Beat Bar Charts: Count Data
03 May 2017 | by Carmen Chan

Pictographs are often subject to ridicule. They are seen to compromise interpretability in favor of glitz. They are only “for kids”: an undeserved reputation which neglects situations where pictographs outperform bar charts. One such situation is for…

Continue reading

Ranking Plot: Main Cause of Death by Age Ranking Plots: Illustrating Data with Different Magnitudes
02 May 2017 | by Tim Bock

Ranking plots, also known as rank flow plots, are particularly useful when comparing data that is complex or that differs in magnitude such as main cause of death by age.

Continue reading

correspondence Analysis Visualization 5 Ways to Visualize Relative Importance Scores from Key Driver Analysis
26 April 2017 | by Tim Bock

Key driver analysis techniques, such as Shapley Value, Kruskal Analysis, and Relative Weights, are useful for working out the most important predictor variables for some outcome of interest (e.g., the drivers of satisfaction or NPS). But,…

Continue reading

Scatterplots A New R Package for Labeled Scatterplots and Bubble Charts
26 April 2017 | by Tim Bock

  The rhtmlLabeledScatter package on github attempts to solve three chronic challenges with labeled scatterplots: readability with large numbers of labels, bubbles, and the use of images.   Four tools for dealing with overlapping labels 1….

Continue reading

heatmap illusion Too Hot to Handle? The Problem with Heatmaps
13 April 2017 | by Tim Bock

Heatmaps are cool. Most people like them. They are so much prettier than a bar chart. The one below, created in Making your data hot: heatmaps for the display of large tables, is both nice to…

Continue reading

Keep updated with the latest in data science.