Effective Data Visualisation

class: center, middle, inverse, title-slide

.title[
# Effective Data Visualisation
]
.subtitle[
## <br><br> Introduction to Data Science
]
.author[
### University of Edinburgh
]
.date[
### <br> 2024/2025
]

---

## Let's have a discussion

.pull-left[
<img src="img/pie-3d.jpg" width="100%" />
]
.pull-right[
<img src="w05-L10_files/figure-html/pie-to-bar-1.png" width="80%" style="display: block; margin: auto;" />
]

---

##  Let's have a discussion

.pull-left[
<img src="w05-L10_files/figure-html/unnamed-chunk-1-1.png" width="100%" style="display: block; margin: auto;" />
]

.pull-right[
<img src="w05-L10_files/figure-html/unnamed-chunk-2-1.png" width="100%" style="display: block; margin: auto;" />
]

---

## Let's have a discussion

.footnote[
Credit: Angela Zoss and Eric Monson, Duke DVS
]

---

## Principles	for	effective	visualizations

- Think about the four respects:
  - Data
  - Mathematics
  - People
  - Computers

- Logical order for a categorical variable
- Ensure everything is readable (long categories names on the y-axis)
- Are there other important variables?
- Select clear and meaningful colours
- Use meaningful title and labels
- Be selective over redundant information

---

## Data

In September 2019, YouGov survey asked 1,639 GB adults the following question:

.pull-left[

> In hindsight, do you think Britain was right/wrong to vote to leave EU?
>
>- Right to leave  
>- Wrong to leave  
>- Don't know

.small[ 
Source: [YouGov Survey Results](https://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/x0msmggx08/YouGov%20-%20Brexit%20and%202019%20election.pdf), retrieved Oct 7, 2019
]
]

.pull-left[

``` r
brexit <- read_csv("data/brexit.csv")
brexit %>% head(n = 2)
```

```
## # A tibble: 2 × 2
##   opinion region       
##   <chr>   <chr>        
## 1 Right   rest_of_south
## 2 Wrong   north
```

``` r
brexit %>% tail(n = 2)
```

```
## # A tibble: 2 × 2
##   opinion region       
##   <chr>   <chr>        
## 1 Wrong   rest_of_south
## 2 Wrong   scot
```
]

---

## Visualise the data

.panelset[
.panel[.panel-name[Plots]

.pull-left[
<img src="w05-L10_files/figure-html/unnamed-chunk-5-1.png" width="80%" style="display: block; margin: auto;" />
]
.pull-right[
<img src="w05-L10_files/figure-html/unnamed-chunk-6-1.png" width="80%" style="display: block; margin: auto;" />

]
]

.panel[.panel-name[Code]
.pull-left[

``` r
ggplot(brexit, aes(x = opinion)) +
  geom_bar()
```
]
.pull-right[

``` r
ggplot(brexit, aes(x = region)) +
  geom_bar()
```
]
]
]

---

## Question

.pull-left[
<br>
.question[
Are people's opinion about Brexit the same or different for each region?
]

``` r
ggplot(brexit, aes(x = region, 
                   fill = opinion)) +
  geom_bar()
```
]

.pull-right[
<img src="w05-L10_files/figure-html/unnamed-chunk-7-1.png" width="80%" style="display: block; margin: auto;" />
]

Are we done? Can we do any better?

---

## Ensure text is readable

.panelset[
.panel[.panel-name[Output]
<img src="w05-L10_files/figure-html/unnamed-chunk-8-1.png" width="60%" style="display: block; margin: auto;" />
]
.panel[.panel-name[Code: Wrangling]

``` r
brexit <- brexit %>%
  mutate(
*   region = fct_recode(
      region,
      London = "london",
      `Rest of South` = "rest_of_south",
      `Midlands / Wales` = "midlands_wales",
      North = "north",
      Scotland = "scot"
    )
  )
```
]
.panel[.panel-name[Code: Visualisation]

``` r
*ggplot(brexit, aes(y = region,
                   fill = opinion)) +
  geom_bar()
```
]
]

---

## Is there a better ording

.panelset[
.panel[.panel-name[Output]
<img src="w05-L10_files/figure-html/unnamed-chunk-9-1.png" width="60%" style="display: block; margin: auto;" />
]
.panel[.panel-name[Code: Wrangling]

``` r
brexit <- brexit %>%
  mutate(
    region = fct_recode(
      region,  
      London = "london",  `Rest of South` = "rest_of_south",
      `Midlands / Wales` = "midlands_wales", North = "north", Scotland = "scot"
    ),
    
*   region = fct_relevel(
*     region,
*     "Scotland", "North", "Midlands / Wales", "Rest of South", "London"
*   ),
  )
```
]
.panel[.panel-name[Code: Visualisation]

``` r
ggplot(brexit, aes(y = region,
                   fill = opinion)) +
  geom_bar()
```
]
]

---

## Faceting rather than stacked

.panelset[
.panel[.panel-name[Output]
<img src="w05-L10_files/figure-html/unnamed-chunk-10-1.png" width="70%" style="display: block; margin: auto;" />
]

.panel[.panel-name[Code: Wrangling]

``` r
brexit <- brexit %>%
  mutate(
    region = fct_recode(
      region,  
      London = "london",  `Rest of South` = "rest_of_south",
      `Midlands / Wales` = "midlands_wales", North = "north", Scotland = "scot"
    ),
    
    region = fct_relevel(   
      region, 
      "Scotland", "North", "Midlands / Wales", "Rest of South", "London"  
    ), 
  )
```
]

.panel[.panel-name[Code: Visualisation]

``` r
*ggplot(brexit, aes(y = opinion,
                   fill = opinion)) +
  geom_bar() +
* facet_grid(~region)
```
]
]

---

## Be selective with redundancy

.panelset[
.panel[.panel-name[Output]
<img src="w05-L10_files/figure-html/unnamed-chunk-11-1.png" width="70%" style="display: block; margin: auto;" />
]

.panel[.panel-name[Code: Wrangling]

``` r
brexit <- brexit %>%
  mutate(
    region = fct_recode(
      region,  
      London = "london",  `Rest of South` = "rest_of_south",
      `Midlands / Wales` = "midlands_wales", North = "north", Scotland = "scot"
    ),
    
    region = fct_relevel(   
      region, 
      "Scotland", "North", "Midlands / Wales", "Rest of South", "London"  
    ), 
  )
```
]

.panel[.panel-name[Code: Visualisation]

``` r
ggplot(brexit, aes(y = opinion,    
                   fill = opinion)) +
  geom_bar() +
  facet_grid(~region) +
  labs(
*   y = NULL
  ) +
* guides(fill = "none")  #Removes legend
```
]
]

---

## Use informative labels

.panelset[
.panel[.panel-name[Output]
<img src="w05-L10_files/figure-html/unnamed-chunk-12-1.png" width="70%" style="display: block; margin: auto;" />
]

.panel[.panel-name[Code: Wrangling]

``` r
brexit <- brexit %>%
  mutate(
    region = fct_recode(
      region,  
      London = "london",  `Rest of South` = "rest_of_south",
      `Midlands / Wales` = "midlands_wales", North = "north", Scotland = "scot"
    ),
    
    region = fct_relevel(   
      region, 
      "Scotland", "North", "Midlands / Wales", "Rest of South", "London"  
    ), 
  )
```
]

.panel[.panel-name[Code: Visualisation]

``` r
ggplot(brexit, aes(y = opinion,    
                   fill = opinion)) +
  geom_bar() +
  facet_grid(~region) +
  labs(
*   title = "Was Britain right or wrong to leave the EU?",
*   subtitle = "YouGov Survey Results, 2-3 September 2019",
*   x = "Frequency",
    y = NULL
  ) +
  guides(fill = "none")
```
]
]

---

## Selection of colours

[colorbrewer2.org](https://colorbrewer2.org/) - Help with selecting a colour pallet

---

## Selection of colours

.panelset[
.panel[.panel-name[Output]
<img src="w05-L10_files/figure-html/unnamed-chunk-14-1.png" width="70%" style="display: block; margin: auto;" />
]

.panel[.panel-name[Code: Wrangling]

``` r
brexit <- brexit %>%
  mutate(
    region = fct_recode(
      region,  
      London = "london",  `Rest of South` = "rest_of_south",
      `Midlands / Wales` = "midlands_wales", North = "north", Scotland = "scot"
    ),
    
    region = fct_relevel(   
      region, 
      "Scotland", "North", "Midlands / Wales", "Rest of South", "London"  
    ), 
  )
```
]

.panel[.panel-name[Code: Visualisation]

``` r
ggplot(brexit, aes(y = opinion,    
                   fill = opinion)) +
  geom_bar() +
  facet_grid(~region) +
  labs(
    title = "Was Britain right or wrong to leave the EU?",  
    subtitle = "YouGov Survey Results, 2-3 September 2019", 
    x = "Frequency",
    y = NULL
  ) +
  guides(fill = "none") +
  scale_fill_manual(values = c(
*   "Wrong" = "#ef8a62",
*   "Right" = "#67a9cf",
*   "Don't know" = "gray"
  ))
```
]
]

---

## Iterate...

.pull-left[
Reminder:
.question[
Are people's opinion about Brexit the same or different for each region?
]
- Has the data visualisation answered this question?
- Is there anything that is unclear?
- What further edits are needed?
- Is this the only visualisation that can answer the question?
- Can you create a better data visualisation?
]

.pull-right[
<img src="w05-L10_files/figure-html/unnamed-chunk-15-1.png" width="100%" style="display: block; margin: auto;" />
]