Category: Visualization

Elizabeth Ricks makes a point:

Today’s post is a tactical one: how to highlight a data point in Excel. 

When we craft visualizations for explanatory purposes—that is, when there’s a specific finding or recommendation that we want to communicate to someone specific—our goal is to drive action. In those cases,  our visuals should emphasize what’s interesting in the data and what requires attention. Highlighting key points in our graph is an important step in creating successful explanatory communications.

Read on for examples as well as how to do this.

Alex Velez notes one additional way that people may interpret your charts differently from your expectations:

Before I share some considerations for presenting data internationally, I want to acknowledge that I am not an expert on different cultures and audiences. In this post, I simply share some of my experiences with the hope that others will provide additional commentary for increased learning. If you have related thoughts, please share in the comments. 

Let’s consider five observations of regional differences I’ve encountered while communicating data.

Read on for those observations.

Mike Cisneros answers a tricky question:

How much context, then, is necessary to include, so that we have an understandable (but un-cluttered) visual, presented with enough background information for the viewer to grasp its meaning, with the key insights and recommended actions emphasized?

When we’ve addressed this question in the past, we’ve relied on the always-true, if sometimes unsatisfying, response of, “It depends.” Every situation is unique, and there’s no checklist or scorecard you can use in every circumstance to ensure that you’ve hit the perfect amount of detail.

Read on for two factors upon which it depends.

Vincent Granville does some marketing:

I explain here how to do it in Excel. But the principle is general. You can produce this type of visualization with other tools. There are many features in Excel that allow you to generate marketing-style pictures. I never really used them, as the result can be cheesy. If overused, you end up with material that looks like advertising from a Casino, or like the “old world wide web”, where blinking fonts and documents with neon colors were popular. But I recently decided to give it a try again, using extreme moderation. I believe my experiment was successful. I will leave it to the reader to have a final say about it.

It turns out a lot less gaudy than I originally imagined.

Amy Esselman explains what a boxplot is:

The “box” part of a boxplot outlines the lower and upper quartiles. Inside the box is a line that indicates the median value. There are lines that extend outside the box—known as the whiskers—to depict the range of values in a given dataset. If there are outliers, then individual dots in line with the whiskers are plotted to denote the extreme values. 

Click through for a depiction of the plot as well as several alternative depictions which can include more information at the cost of added complexity.