# Module 3.1: Figures gone wrong (or a similar title)

(Outline for 3.1 in [here](https://github.com/alan-turing-institute/rds-course/issues/20))

## Introduction

<!-- …(Warning to Callum: All this text is kind of a brain dump, so the writing is not as good as it should be).-->

In this module we aim to provide useful tips that can improve the way you use visualisation to effectively and accurately comunicate insights from data.

- [name=Camila] *What I think we should write here*:
- Mention importance of data viz as tool for storytelling and outreach, specially how this has become fundamental in the context of the pandemic.

This module is structured as follows: first we will discuss the basics of data visualisation (rules of the game), from here we will look at a variety of graphs and discuss which kind of graphs are better for different type of scenarios. Then we will follow with a discussion of how to tell a story with data visualisation and how to include context and ethical considerations in conceiving the story to tell. Finnaly we will talk about using data visualisation to better understand your own data.

**References**:

This module is largely based on the [Fundamentals of Data Visualization](https://clauswilke.com/dataviz/) book by Claus O. Wilke but we also use resources from [the visualising data site](https://www.visualisingdata.com/) and other pages that we will be referencing as we go.

### Figures gone wrong

- [name=Camila] *What I think we should write here*:
- Mention how we all know how to make plots, and probably have done thousands on our professional lifetimes, however this is not necesarily an easy job and there are plenty of examples of figures that have gone wrong in different ways, for example:

<!-- …(Note to Callum: I managed to group these in 4 groups: misleading information, lack of uncertainty, overcomplicated figures and problems with anotations. Maybe we can come up with better names for the groups).-->

#### **Misleading information**

- Axes does not start at zero and exaggerates the effect (from [here](https://www.callingbullshit.org/tools/tools_misleading_axes.html)):


- Purposely hide information to exagerate an idea:

([from here](https://www.visualisingdata.com/2015/10/if-your-visuals-deceive-your-message-deceives/))
([good version](https://twitter.com/emschuch/status/649690759453646848))

- Axis scale changing on the figure

- Failiure to normalise:

#### Figures that fail to represent uncertainty

- Beware of averages (from Factfulness book)!


#### **Over complicated figures**:

- Over complicated figure (a bar chart would have been enough):


[source](https://badvisualisations.tumblr.com/post/184827953341/this-is-not-all-you-need-to-click-through-to-this)

- Too much information:
[source](https://twitter.com/10DowningStreet/status/1322614557181960195)

#### **Problems with anotations**

- No axis:


[source](https://twitter.com/Rodpac/status/1250764503861600256?s=20)

- Duplicated labeling


from this [page](https://www.visualisingdata.com/)

- Ugly anotations


[source](https://badvisualisations.tumblr.com/post/184827953341/this-is-not-all-you-need-to-click-through-to-this)