# Do you know what is the right connection in a Dot Plot?

## Incremental Improvements #05: Dot Plot

May 12 · 4 min read

Last week I came across some unusual dot plot that was done by The Economist. At first, I thought that this is a refreshing way of using this chart. But the longer I look at it the stranger it seemed. The chart presents the feelings about removal of US troops from Afghanistan. It was created to allow the comparison of three “Yes/No” questions among three different groups — all Americans, republicans, and democrats.

My problem with this chart is its overcomplication and misleading usage of the line. In the dot plot, the length of the line shows the difference between two data points — and the longer it is, the bigger the difference.

In this chart is quite the opposite, a shorter line means a more polarized opinion on the subject. But even this is not always true because the chart doesn’t show all answers (have you noticed that the numbers don’t sum up to 100%?), so imagine two situations in which the dots are close to each other: 1) the two dots are close to each other in the middle of the scale, which means that around 50% of people agree and another 50% disagree, so the opinions are polarized; 2) the two dots are close to each other around the 20-percent-point, which means that the majority gave a neutral answer or no answer at all.

You can see that this way of presenting data is very confusing. So how would I change it?

# Good data visualization solutions✔️

Elements that work in this chart?

• Chart selection — dot plot is a good chart selection because it allows comparison between two groups (or even three if you add an overall result) or two timestamps (before and after comparison). I would only change the way that the points are connected.
• Comparing different questions — using the same scale and showing questions together show a fuller picture. The audience can easily compare the diversity of views among different groups (democrats, republicans, and overall) and within different aspects (comparison across the questions).

Elements that don’t work in this chart?

• Way of encoding positive and negative — in the original chart the opposing answers for each question are encoded with color — red means good (approve / yes) and blue means bad (disapprove / no). This is counterintuitive (naturally we associate red with warning / danger / negative) and hinders the comparison. This is especially noticeable when the answers “change their places”, like in the second question. Were you able to get the right insight quickly?
• Connecting opposite answers —which brings confusion to how to interpret the length of the line.

# How to make this data visualization better❔

Step-by-step improvements

• Show positive and negative answers on opposite sides and remove the line — so the answers are encoded by position (left / right side of the zero point), not with the color. This simplifies the chart and allows for convenient cross-group comparison.
• Assign a color to the groups — and put all groups in one row.
• Connect the same answers with the line — so the length of the line will encode deviation among the groups.
• Provide context for easier comparison— by adding the labels. Because we are comparing diversification among groups, the important information is the overall result and percentage difference.
• Add formatting — using darker shade makes a zero point more visible and comparison easier.

Do you think it delivers the message better? Would you improve something else? If you have some suggestions for the next makeover feel free to drop them in the comments.

This is a series dedicated to small changes that can be done in order to make a visual design better. In each post, I analyze the data visualization to see what works, what doesn’t, and what to do to improve it.

Previous chart makeovers: small multiples chart, bar chart usage, stacked chart alternative, and Likert scale.

If you are looking for some dataviz inspirations check the essential resources every data visualization designer needs to know

## Analytics Vidhya

Analytics Vidhya is a community of Analytics and Data…

Written by

## Weronika Gawarska-Tywonek

Data Visualization Designer | Tableau Associate | Sociologist with passion for aesthetics

## Analytics Vidhya

Analytics Vidhya is a community of Analytics and Data Science professionals. We are building the next-gen data science ecosystem https://www.analyticsvidhya.com

Written by

## Weronika Gawarska-Tywonek

Data Visualization Designer | Tableau Associate | Sociologist with passion for aesthetics

## Analytics Vidhya

Analytics Vidhya is a community of Analytics and Data Science professionals. We are building the next-gen data science ecosystem https://www.analyticsvidhya.com

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app