Statistics and Quantitative Methods (S1)

.title[
# Statistics and Quantitative Methods (S1)
]
.subtitle[
## Week 2 — Workshop
]
.author[
### Dr Stefano Coretta
]
.institute[
### University of Edinburgh
]
.date[
### 2022/09/27
]

---

# Update the sqmf package

Time to update the sqmf package!

Run the following in the console:

``` r
remotes::install_github("stefanocoretta/sqmf")
```

---

# Data visualisation

![](../../img/data-quant.png)

---

# Graphic systems

- Base R.

- lattice.

- ggplot2.

- more...

---

# But before that... R SCRIPTS!

1. **Create a folder** inside your RStudio project named `code`.

- You can do this from within RStudio, in the Files panel, or through the OS Finder/File explorer.

2. **Create a new R Script**.

3. Save the R script as `week_2.R` in `code/`.

---

# R script basics

- Write each command into **its own line**.

- Commands should be written in the **order they must be executed**. In other words, later lines of code depend on earlier lines of code, never the opposite.
  
- To **execute a command**, place the text cursor (or caret) anywhere in the line you want to execute and press `CMD/CTRL+ENTER`.

- You can **add comments** in a scripts by starting a line with `#`.

---

# Base R plotting function

```r
x <- 1:10
y <- x^3
plot(x, y)
```

---

# Base R plotting function

---

# Base R plotting function

```r
x <- seq(1, 10, by = 0.01)
y <- x^3
plot(x, y, type = "l", col = "purple", lwd = 3, lty = "dashed")
```

---

# Base R plotting function

---

# Packages to the rescue

<br>

- The R **library** contains the packages you have installed.

- Base R.

- Extra packages.

- Install extra packages with `install.packages()`.

.bg-washed-blue.b--dark-blue.ba.bw2.br3.shadow-5.ph4.mt5[

Run this in the console:

```r
install.packages("fortunes")
```

]

---

# Packages to the rescue

To use a package installed in the library, you `attach` the package with `library()`.

.bg-washed-blue.b--dark-blue.ba.bw2.br3.shadow-5.ph4.mt5[

Run this in the console:

```r
library(fortunes)
```

]

---

# Packages to the rescue

Now you can use the functions provided by the attached package

.bg-washed-blue.b--dark-blue.ba.bw2.br3.shadow-5.ph4.mt5[

Run this in the console:

```r
fortune()
```

```
## 
## However, if you want to do this at all efficiently for a data frame, start with
## my solution not Uwe's [...]
##    -- Brian D. Ripley (in a second reply to a question related to data frame
##       manipulations)
##       R-help (June 2004)
```

]

---

# R script basics

- Go ahead and add the code to attach the following packages in your R script:

- **tidyverse**
      - This is a "meta-package", a packages of packages. Among these there is ggplot2.
  - **glottologR**
      - This is a data package, a package that provides users with data. This package contains data from Glottolog.

- You can now attach the endangerment data with `data("glot_aes")`.

The top of your script should look like this (also see next slide for a more convoluted example).

``` r
library(tidyverse)
library(glottologR)
data("glot_aes")
```

???

The tidyverse and glottologR packages should have been installed automatically when you installed sqmf.

If for any reason that didn't work, run the following lines in the console:

``` r
install.packages("tidyverse")
remotes::install_github("stefanocoretta/glottologR")
```

---

# R script basics

---

To check what the data looks like, just call the data in the console, and you will see the first few lines of the data table.

```r
glot_aes
```

```
## # A tibble: 8,345 × 18
##    ID           Langu…¹ Param…² Value Code_ID Comment Source codeR…³ AES   Name 
##    <chr>        <chr>   <chr>   <chr> <chr>   <chr>   <chr>  <lgl>   <fct> <chr>
##  1 kolp1236-aes kolp12… aes     3     aes-sh… Kol (1… hh:he… NA      shif… Kol …
##  2 tana1288-aes tana12… aes     3     aes-sh… Tanahm… hh:he… NA      shif… Tana…
##  3 touo1238-aes touo12… aes     3     aes-sh… Touo (… hh:he… NA      shif… Touo 
##  4 bert1248-aes bert12… aes     3     aes-sh… Fadash… hh:he… NA      shif… Berta
##  5 sius1254-aes sius12… aes     6     aes-ex… Siusla… hh:he… NA      exti… Sius…
##  6 cent2045-aes cent20… aes     6     aes-ex… Jalaa … <NA>   NA      exti… Jalaa
##  7 else1239-aes else12… aes     3     aes-sh… Elseng… hh:he… NA      shif… Else…
##  8 taia1239-aes taia12… aes     4     aes-mo… Taiap … hh:he… NA      mori… Taiap
##  9 pyuu1245-aes pyuu12… aes     3     aes-sh… Pyu (4… hh:he… NA      shif… Pyu  
## 10 mato1253-aes mato12… aes     6     aes-ex… Arára … hh:he… NA      exti… Mato…
## # … with 8,335 more rows, 8 more variables: Macroarea <chr>, Latitude <dbl>,
## #   Longitude <dbl>, Glottocode <chr>, ISO639P3code <chr>, Countries <chr>,
## #   Family_ID <chr>, Language_ID.y <chr>, and abbreviated variable names
## #   ¹Language_ID, ²Parameter_ID, ³codeReference
```

---

Alternatively, you can also view the data as a table in RStudio:

- By clicking on the name of the data in the Environment panel.

- Or by running the following in the console:

``` r
view(glot_aes)
```

---

# Language endangerment

---

```r
ggplot(
  data = glot_aes,
  mapping = aes(x = AES)
) +
  geom_bar()
```

- `ggplot()` function.

- Arguments:
  - `data` to specify which data to plot.
  - `mapping` to specify mapping between columns in the data and "aesthetics".
  
- `geom_bar()` to add a bar geometry.

---

---

```r
ggplot(
  data = glot_aes,
  mapping = aes(x = AES)
)
```

]

![](index_files/figure-html/aes-plot-1-1.png)

]

---

```r
ggplot(
  data = glot_aes,
  mapping = aes(x = AES)
) +
  geom_bar()
```

]

![](index_files/figure-html/aes-plot-2-1.png)

]

---

```r
ggplot(
  data = glot_aes,
  mapping = aes(x = AES, fill = AES)
) +
  geom_bar()
```

]

![](index_files/figure-html/aes-plot-3-1.png)

]

---

```r
ggplot(
  data = glot_aes,
  mapping = aes(x = Macroarea)
) +
  geom_bar()
```

]

![](index_files/figure-html/aes-plot-4-1.png)

]

---

```r
ggplot(
  data = glot_aes,
  mapping = aes(x = Macroarea, fill = AES)
) +
  geom_bar()
```

]

![](index_files/figure-html/aes-plot-5-1.png)

]

---

```r
ggplot(
  data = glot_aes,
  mapping = aes(x = Macroarea, fill = AES)
) +
  geom_bar(position = "dodge")
```

]

![](index_files/figure-html/aes-plot-6-1.png)

]

---

# Politeness and f0

---

Let's play around with other data.

- When you installed the sqmf package, you also installed some data tables with it.

- To attach the data from the sqmg package you have first to attach the package.

- Add the following line of code at the top of your script, after the other `library()` lines:

``` r
library(sqmf)
```

- Now check the list of data tables available in the package:

``` r
data(package = "sqmf")
```

- Finally, attach the `polite` data table.

``` r
data("polite")
```

---

```r
ggplot(
  data = polite,
  mapping = aes(x = attitude, y = f0mn)
) +
  geom_point()
```
]

---

```r
ggplot(
  data = polite,
  mapping = aes(x = attitude, y = f0mn)
) +
* geom_jitter()
```
]

---

```r
ggplot(
  data = polite,
  mapping = aes(x = attitude, y = f0mn)
) +
* geom_jitter(width = 0.2)
```
]

---

```r
ggplot(
  data = polite,
  mapping = aes(x = attitude, y = f0mn)
) +
  geom_jitter(width = 0.2) +
* facet_grid(~ gender)
```
]

---

```r
ggplot(
  data = polite,
  mapping = aes(x = attitude, y = f0mn)
) +
  geom_jitter(width = 0.1) +
* facet_grid(musicstudent ~ gender)
```
]

---

# Infant gestures

---

Now, let's use the `gestures` data.

```r
# This data table is in the sqmf package.
data("gestures")
gestures
```

```
## # A tibble: 1,620 × 11
##    dyad  backgro…¹ months task  gesture count…² count ct_raw    ct pro_r…³    id
##    <chr> <chr>      <dbl> <chr> <chr>     <dbl> <dbl>  <dbl> <dbl> <chr>   <dbl>
##  1 b01   Bengali       10 five  reach         5     5      1     1 no          3
##  2 b01   Bengali       10 five  point         0     0      0     0 no          2
##  3 b01   Bengali       10 five  ho_gv         0     0      0     0 no          1
##  4 b01   Bengali       10 tp1   reach         0     0      0     0 no          6
##  5 b01   Bengali       10 tp1   point         0     0      0     0 no          5
##  6 b01   Bengali       10 tp1   ho_gv         0     0      0     0 no          4
##  7 b01   Bengali       10 tp2   reach         0     0      0     0 no          9
##  8 b01   Bengali       10 tp2   point         0     0      0     0 no          8
##  9 b01   Bengali       10 tp2   ho_gv         0     0      0     0 no          7
## 10 b01   Bengali       11 five  reach         7     8      3     3 yes         3
## # … with 1,610 more rows, and abbreviated variable names ¹background,
## #   ²count_raw, ³pro_rata
```

---

```r
gestures %>%
  ggplot(aes(months, count, group = id)) +
  geom_line(alpha = 0.5) +
  geom_point()
```
]

---

```r
gestures %>%
  ggplot(aes(months, count)) +
  geom_point()
```
]

---

```r
gestures %>%
  ggplot(aes(months, count)) +
  geom_line(alpha = 0.2) +
  geom_point()
```
]

---

```r
gestures %>%
* ggplot(aes(months, count, group = id)) +
  geom_line(alpha = 0.2) +
  geom_point()
```
]

---

# Lexical decision task

---

Data from MALD.

```r
data("mald_1_1")
mald_1_1
```

```
## # A tibble: 5,000 × 7
##    Subject Item        IsWord PhonLev    RT ACC     RT_log
##    <chr>   <chr>       <fct>    <dbl> <int> <fct>    <dbl>
##  1 15308   acreage     TRUE      6.01   617 correct   6.42
##  2 15308   maxraezaxr  FALSE     6.78  1198 correct   7.09
##  3 15308   prognosis   TRUE      8.14   954 correct   6.86
##  4 15308   giggles     TRUE      6.22   579 correct   6.36
##  5 15308   baazh       FALSE     6.13  1011 correct   6.92
##  6 15308   unflagging  TRUE      7.66  1402 correct   7.25
##  7 15308   ihnpaykaxrz FALSE     7.47  1059 correct   6.97
##  8 15308   hawk        TRUE      6.09   739 correct   6.61
##  9 15308   assessing   TRUE      6.37   789 correct   6.67
## 10 15308   mehlaxl     FALSE     5.80   926 correct   6.83
## # … with 4,990 more rows
```

---

```r
mald_1_1 %>%
  ggplot(aes(RT)) +
  geom_density() +
  geom_rug()
```
]

---

```r
mald_1_1 %>%
* ggplot(aes(RT, colour = IsWord)) +
  geom_density() +
  geom_rug()
```
]

---

```r
mald_1_1 %>%
* ggplot(aes(RT, fill = IsWord)) +
  geom_density(alpha = 0.2) +
  geom_rug()
```
]

---

```r
mald_1_1 %>%
* ggplot(aes(RT, fill = ACC)) +
  geom_density(alpha = 0.2) +
  geom_rug() +
  facet_grid(~ IsWord)
```
]

---

???

To run the tutorial, go to the Tutorial tab in the top right panel of RStudio. You will find the tutorial at the bottom of the list.

Or, run the following:

``` r
learnr::run_tutorial("02_dataviz", "sqmf")
```

---

# Further resources

- For a detailed overview with exercises of ggplot2, see Chapter 3 of the [R4DS book](https://r4ds.had.co.nz/data-visualisation.html).

- [Fundamentals of Data Visualisation](https://clauswilke.com/dataviz/).

- [Directory of visualisations](https://clauswilke.com/dataviz/directory-of-visualizations.html).
- [Data viz catalogue](https://datavizcatalogue.com/index.html).
- [Data Viz project](https://datavizproject.com).
- [Top 50](http://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html).
- [Data Viz](https://datavizm20.classes.andrewheiss.com/).

**Tutorials**

- [Raincloud plots](https://wellcomeopenresearch.org/articles/4-63).
- [Labels](https://www.cararthompson.com/talks/user2022/).
- [Graphic design](https://rstudio-conf-2022.github.io/ggplot2-graphic-design/).
]

- [Use colour wisely](https://albert-rapp.de/post/2022-02-19-ggplot2-color-tips-from-datawrapper/).
- [ColorBrewer2](https://colorbrewer2.org/#type=sequential&scheme=BuGn&n=3).
- [MetBrewer](https://github.com/BlakeRMills/MetBrewer).

**Recommendations**

- [Same stats different data](https://www.autodesk.com/research/publications/same-stats-different-graphs).
- [I stopped using box plots](https://nightingaledvs.com/ive-stopped-using-box-plots-should-you/).
- [Issues with error bars](https://www.data-to-viz.com/caveat/error_bar.html).
- [Behind bars](https://stats.stackexchange.com/a/367889).
- [Visual Word Paradigm](https://link.springer.com/article/10.3758/s13423-022-02143-8).
]