Week 44: Everybody poops

For week 44 we asked our community to tackle a clean dataset about a dirty matter: washing your hands after pooping. As it turns out, a lot of people in the UK don’t always wash their hands after using the bathroom and we wanted to see what kind of different visualizations you all would come up with.

Thanks for #TC18 and many new people joining the project, we enjoyed seeing a lot of submissions this week.

If you’re new to #MakeoverMonday, we would like to welcome you to the project and to our community. It’s great to have you here and we look forward to seeing your vizzes, watching you learn and helping you along the way.

In this weekly recap blog we look back at the week, tackle two lessons – one being about analysis, the other about design – and highlight our favorite vizzes, which capture the data and the messages in a really effective way. While we include inspirational designs, the purpose of the favorites section is to showcase best practices and help people see good ways to represent data in an effective way, that communicates clearly.

LESSON 1: WHAT IS YOUR VIZ ABOUT?

This week during Viz Review I noticed a few vizzes which talked about hand-washing and that some people don’t always wash their hands. And that was it. No reference to bathroom visits, poo or anything else related. Which makes those titles wrong.

Saying ‘xx% of men don’t always wash their hands’ and citing the YouGov survey as the source is incorrect when there is no reference whatsover to the actual topic, which is hand-washing after going to the toilet.

It is important to provide your audience with enough context. Always assume that someone who looks at your viz has never seen the original visualization, doesn’t know the data and hasn’t read the article or surrounding information. Giving the people who look at your work enough information in the title, subtitle or annotations and labels is important, especially if your viz does not make the topic and conclusions very obvious.

Kate Brown makes a clear statement in the subtitle of her viz to help her audience understand quickly what the viz is about.

As you build your next vizzes, focus on having a clear title and/or explanatory subtitle to make sure your viz isn’t misunderstood. This week we visualized data about handwashing after using the bathroom, not handwashing before making food, or after gardening or in any other situation.

The less ambiguous you can make your viz, the better.

LESSON 2: AVERAGING AVERAGES

With survey data, we typically don’t have the same number of respondents in each category, so we cannot simply average their results, i.e. take the % for men and the % for women, add them together and divide by 2 to have the average number.

For example, if you have a survey with responses from 10 men and 1000 women, calculating the average of their responses this way, the weightings are off, because they don’t take the sample size into account.

If you want to calculate the overall average, you must take into account sample sizes. If you don’t have sample sizes, it’s best to stay clear of the overall average. Better not to include it than including a wrong number.

For more information about tackling survey data, head to Steve Wexler’s website and check out the resources he has created over time.

FAVORITES

Author: Lindsay Betzendahl
Link: Tableau Public

Author: Jennie Cotton
Link: Tableau Public

Author: Rodrigo Calloni
Link: Tableau Public