A reProducible woRkflow with Quarto

<style>

.center2 {
  margin: 0;
  position: absolute;
  top: 50%;
  left: 50%;
  -ms-transform: translate(-50%, -50%);
  transform: translate(-50%, -50%);
}

.rcorners1 {
 margin: auto;
 border-radius: 25px;
 background: #ada500;
 padding: 10px;
# width: 50%;
}
</style>

.remark-code, .remark-inline-code { font-family: 'Source Code Pro', 'Lucida Console', Monaco, monospace;
                                    font-size: 90%;
                                  }

</style>

# A reProducible woRkflow with Quarto
.font160[
.SW-greenD[Part 1]
]
.font120[
.SW-greenD[*About reproducibility and R*]
]
Sven De Maeyer & Tine van Daal

---
class: inverse-green, left

# Overzicht

.center2[
1. RepRoducibility?! --- ([click here](#part1))
2. Sta`R`t to be `R`eproducible --- ([click here](#part2))
3. A woRld of packages! --- ([click here](#part3))
4. Working with `R`-projects --- ([click here](#part4))
5. Importing data --- ([click here](#part5))

]

---
class: inverse-green, center, middle
name: part1

# 1. Rep`R`oducibility?!

---

## Key principles of Open Science

--

.font140[.SW-greenD[**Transparency**]] 
.SW-greenL[Openness about our methodology, data, and results.]

.font140[.SW-greenD[**Openness**]] .SW-greenL[Research and research outputs are freely available to everyone.]

.font140[.SW-greenD[**Reproducibility**]] .SW-greenL[Research methods and data are shared in such a way that others can **follow, reproduce, and reuse** the decisions we have made.]

.font80[Definitions taken from Edubron's [mission statement](https://www.uantwerpen.be/en/research-groups/edubron/research/open-science/) regarding Open Science.]

---

## A reproducible workflow = your external (research) hard drive

.Large["*Can you redo that analysis to check ...?*"]

--
<center>
<img src="img/reproducible_not.jpg" width="600" height="453" />
</center>

---

## Reproducible or replicable?

.footnote[
.font50[
 * Figure from The Turing Way Community & Scriberia (2024) downloaded from https://zenodo.org/records/13882307 *
]
]
]

--
.pull-right[
 
 
 
.right[
.font140[
*What does it take to do reproducible research?*
]
]
]

---

## Towards a reproducible workflow

---

## A reproducible quantitative workflow

.footnote[
.font50[
 * Wickham, H., Cetinkayak-Rundel, M., & Grolemund, G. (2023) R for Data Science (second edition). O'Reilly. Also available online: https://www.edureka.co/blog/r-programming-language *
]
]

---

## Why we talk about rep<img src="img/Rlogo.svg" width="5%"/>oducibility?
 
.pull-left[
`R` is a powerful tool for:
- Data cleaning and preparation
- Statistical analysis
- Data visualisation
- ...

]

.pull-right[
 
`R` facilitates .SW-greenD[**reproducibility**] (and transparency) by allowing you to write a **script**.
 
 
 
 
 
.font80[

``` r
library(tidyverse)              #load tidyverse package
library(datasauRus)             #load datasauRus package
datasaurus_dozen %>%
  filter(dataset=="star") %>%   #filter rows
  ggplot(
    aes(x = x, y = y)           #define x- and y-axis
    )+
  geom_point(size = 1.2) +      #change size of points
  annotate(                     #add text
    geom = "text", 
    x = 60, y = 52, size = 8, fontface = "bold",
    color = "#AAAA17",
    label = "R is an open science staR!"
    ) +
  theme_void() +                #set theme of plot
  theme(legend.position = "none")
```
]
]

---
class: inverse-green, center, middle
name: part2

# 2. Sta`R`t to be `R`eproducible

---

## RStudio to make you`R` life easie`R`
<left>
<img src="img/rstudio-panes-labeled.jpeg" width="55%"/>
</left>
 
 
.footnote[
.font50[
 * Figure from RStudio User Guide (Release 2026.01.01) https://docs.posit.co/ide/user/ide/get-started/ *
]
]

---
class: inverse-blue, center, middle

## Let's make a script in `RStudio`
.rcorners1[.Large[
 * Time to get started with the real work and code together*
]
.small[You can find the code on the next slides.

There is also a script `Fruit.R` that contains the same code (see course material for Part 1 online)]
]

---

### .SW-greenD[Code-Blocks: character vector]

We create a vector and call it _.UA-blue[Fruit]_  
.small[

``` r
Fruit <- c("Apples", "Bananas", "Lemons", "Berries", "Peaches", NA)
```
]
.font80[
- .UA-red[`c()`] represents .SW-greenD[*concatenate*], all elements between the brackets ( separated by `,`) are 'merged' into one element

- .UA-red[`<-`] means that we want to store the result in an object (a vector) that we call `Fruit`

- the .UA-red[`" "`] that surround the elements indicate that these are of the type *character*]

Let's look at the object _.UA-blue[Fruit]_ by printing it to the console  
.small[

``` r
Fruit
```

```
## [1] "Apples"  "Bananas" "Lemons"  "Berries" "Peaches" NA
```
]
We can also check the structure of the object _.UA-blue[Fruit]_ by using the **.UA-red[`str()`]** function
.small[

``` r
str(Fruit)
```

```
##  chr [1:6] "Apples" "Bananas" "Lemons" "Berries" "Peaches" NA
```
]

---
### .SW-greenD[Code-Blocks: numeric vector]

We create a vector called _.UA-blue[Weight]_  
.small[

``` r
Weight <- c(230, 191, 93, 100, 48, 244)
```
] 
 
In this case, the elements are 'recorded' as numeric elements 
.small[

``` r
str(Weight)
```

```
## num [1:6] 230 191 93 100 48 244
```
]
 
Now, we can start calculating... For example, we can apply the function .UA-red[`mean( )`] 
.small[

``` r
mean(Weight)
```

```
## [1] 151
```
]

---

### .SW-greenD[Code-Blocks: logical vector]

We create a vector called _.UA-blue[Yellow]_

``` r
Yellow <- c(F, T, T, F, F, F)
```
 
The element of the vector _.UA-blue[Yellow]_ are 'logical operators' (TRUE or FALSE)

``` r
str(Yellow)
```

```
## logi [1:6] FALSE TRUE TRUE FALSE FALSE FALSE
```
 
Let us have a closer look at the vector _.UA-blue[Yellow]_

``` r
Yellow
```

```
## [1] FALSE  TRUE  TRUE FALSE FALSE FALSE
```

---
### .SW-greenD[Code-Blocks: a data frame]
.pull-left[
We create a data.frame called _.UA-blue[Fruit_data]_ using the function **.UA-red[`data.frame( )`]** 
.small[

``` r
Fruit_data <- data.frame(Fruit, Weight, Yellow)
```
]
 
Let's have a look at *.UA-blue[Fruit_data]*
.small[

``` r
Fruit_data
```

```
## Fruit Weight Yellow
## 1 Apples 230 FALSE
## 2 Bananas 191 TRUE
## 3 Lemons 93 TRUE
## 4 Berries 100 FALSE
## 5 Peaches 48 FALSE
## 6 <NA> 244 FALSE
```
]
]
.pull-right[
We can check the structure of the object _.UA-blue[Fruit_data]_ 
.small[

``` r
str(Fruit_data)
```

```
## 'data.frame':	6 obs. of  3 variables:
##  $ Fruit : chr  "Apples" "Bananas" "Lemons" "Berries" ...
##  $ Weight: num  230 191 93 100 48 244
##  $ Yellow: logi  FALSE TRUE TRUE FALSE FALSE FALSE
```
]
]

---
### .SW-greenD[Code-Blocks: subsetting a data frame]

``` r
Fruit_data$Weight
```

```
## [1] 230 191  93 100  48 244
```
]
By **indexing** specific columns- (and/or) row-numbers can be selected .UA-red[`[row, column]`]
]

.pull-right[
**Examples of indexing**: 
- Retrieve element that is located at row 1 and column 1    
.small[

``` r
Fruit_data[1,1]
```

```
## [1] "Apples"
```
]

- Retrieve all elements in column 3  
.small[

``` r
Fruit_data[,3]
```

```
## [1] FALSE  TRUE  TRUE FALSE FALSE FALSE
```
]

- Retrieve all elements in row 3  
.small[

``` r
Fruit_data[3,]
```

```
##    Fruit Weight Yellow
## 3 Lemons     93   TRUE
```
]
]

---

## Type of objects

- matrices

- arrays

- *data frames*

- lists

- *functions*

...
]

<img src="img/Data_Types.png" width="80%" height="80%" />
]

---

## Vectors

---

## This is a *repRoducible* workshop!

Various types of files used during the workshop:

- R-scripts (as example) (.R)

- Rmd-files (for slides) (.Rmd)

- Qmd-files (for exercises) (.Qmd)

- data sets (.sav, .csv, ...)

- html-version of slides and exercises (.html)

All these files can be found at the dedicated web-page: https://r-workshop-edubron26.netlify.app/

---
class: inverse-green, center, middle
name: part3

# 3. A woRld of packages!

---

## R-Packages?!

- Package = extension of the `Base`-functions

- Range from .SW-greenD[specialised] to .SW-greenD[generic/universal] packages

- Overview on .font100[https://cran.r-project.org/web/packages/available_packages_by_name.html]

- Which package(s) to use?
]

.pull-right[
<img src="img/R_packages.jpg" width="100%" height="100%" style="display: block; margin: auto 0 auto auto;" />
]

---

## The `tidyverse` package is a .UA-red[**MUST-HAVE**] for everyone!
 
 
<img src="img/Tidyverse_screen.jpg" width="70%" height="70%" />

---

## A reproducible workflow with tidyverse

<center>
<img src="img/tidyverse-package-workflow.png" width="75%"/>
</center>
 
 
.footnote[
.font50[
 * Figure from Teach data Science https://teachdatascience.com/tidyverse/ *
]
]

---

## .UA-red[ `install.packages( )`] and .SW-greenD[`library( )`]

- by 'clicking' in `RStudio`

- **by using a function in the console or script**

*What Sven usually does: `install.packages()`*   
*What Tine usually does: click and go!*

For example,

``` r
install.packages("tidyverse", dependencies = T)
```
]

- by 'clicking' in `RStudio`

- **by using a function in de console or script**

*What we both usually do: `library()`*

For example,

``` r
library("tidyverse")
```
]

---

# 4. Working with `R`-projects

---

## Handling all these files
 
During analyses, we will regularly **handle different files**:

- data sets
- scripts
- output (html-files; figures; ...)

Consequently, we should **refer to these files in our code**.

---

## Taking the right path ...

There are two ways to refer to these files.

We can do this using an *ABSOLUTE path*.
Below is an example  of an absolute path:

``` r
'c:/Users/Sven/Mijn Documenten/UAntwerpen/Analyses/ProjectX/R_Script/Analysescript1.R' 
```

.font80[Absolute paths make it difficult to share your work; you get into trouble when you change laptops; ...
]
]

To avoid these problems, `RStudio` introduced the concept of an *.UA-red[`R`-Project].*

Within a project you can use *RELATIVE paths* (that starts from the folder in which you save the project).
 
Below is an example of a relative path:

``` r
'~R_Script/Analysescript1.R' 
```
]

---

## Sharing projects? 
 
If Tine puts a project of Sven on her laptop, the .SW-greenD[RELATIVE paths] might get her into trouble. 
For example, the Rmd-file 'Slides_part1.Rmd' is located at

`'c:/Users/Sven/Dropbox/R_workshop_Edubron26/Presentations/Part 1/Slides_part1.Rmd' `

Of course, Tine doesn't have the same structure on her laptop...

To facilitate sharing of projects, the package **`here`** has been created.

---

## Creating relative paths with function .UA-red[`here()`]

The function `**here()**` creates a paths relative to the top-level directory of my laptop.

``` r
library(here)
here()
```

```
## [1] "/Users/svendemaeyer/Library/CloudStorage/OneDrive-UniversiteitAntwerpen/Bestanden van Tine van Daal - A repRoducible woRkflow with Quarto"
```

``` r
here("Presentations", "Part 1", "Slides_part1_Rmd")
```

```
## [1] "/Users/svendemaeyer/Library/CloudStorage/OneDrive-UniversiteitAntwerpen/Bestanden van Tine van Daal - A repRoducible woRkflow with Quarto/Presentations/Part 1/Slides_part1_Rmd"
```

---
class: inverse-blue, center, middle
## Exercises

.small[You can find the folder `Exercises` at the Github: <https://github.com/Sdemaeyer2/R_workshop_Edubron26/>. Download this folder and put it somewhere on your laptop.]
]
]

---

## Creating a project in `RStudio`

Click on .UA-red[`File/New Project...`]

Next, you can choose:

---

# 5. Importing data

---

## Data comes in different formats

Data exist in various formats, but the most common ones are:

- MS Excel
- SPSS / Stata / SAS 
- text (csv, tab-delimited)

Various methods and packages have been developed to import these types of data

We will show some possibilities

---

## Importing MS Excel data with the .UA-red[`readxl`] pakket

Make sure that the package is installed on your laptop
Then, you can use the code below to import the data
.small[

``` r
library(readxl)
Data <- read_excel(
 path = "<path to map and file>"
)
```
]
*Imagine*: you have an excel-file .SW-greenD[**Datacollection1**] with multiple sheets 
*Goal*: you only want to import the sheet with name .SW-greenD[**Group2**] and save it as a data frame .SW-greenD[**Data_Group2**]
.small[

``` r
Data_Groep2 <- read_excel(
 path = "<path to map and file>/Datacollection1.xlsx",
 sheet = "Group1"
)
```
]
You shouldn't use the argument `sheet` if: 
- the excel-file only includes 1 sheet
- there are multiple sheets but you only want to import the first one

---

## Importing SPSS data with package .UA-red[`foreign`]

Make sure that the package is installed on your laptop
Then, you can use the code below to import the data

``` r
library(foreign)
Data <- read.spss(
 file = "<path to map and file>",
 use.value.labels = FALSE,
 to.data.frame = TRUE
)
```

> .SW-greenD[`use.value.labels`] argument: 
In SPSS, data is labelled. Do you want these labels to appear in the R data-frame? 
*For example. Variable with 3 categories: 1 = "Low", 2 = "Average", 3 = "High". These labels are included in the SPSS-file. You can choose to import these labels in the R data frame (`use.value.labels = TRUE`) or only the numbers 1, 2 and 3 that represent these categories (`use.value.labels = FALSE`)*

---

## Importing csv-files with function .UA-red[`read.table( )`]

.pull-left[
csv-files usually look like this .font80[(separated by .UA-red[.large[`,`]] or .UA-red[.large[`;`]]) ]
.small[

``` r
Column1, Column2, Column3
1, 3, 5
2, 4, 6
8, 10, 99
```
]

To import these data, the function .UA-red[`read_table( )`] is most convenient...
.small[

``` r
Data <- read.table(
 file = "<path to map and file>",
 header = TRUE,
 sep = ",", 
 dec = "."
)
```
]
]
.pull-right[
- **.SW-greenD[`header`]** argument: 
The first row includes column (variable) names or not?

- **.SW-greenD[`sep`]** argument: 
Which character is used to separate the values in the different columns?

- **.SW-greenD[`dec`]** argument: 
Which character is used to indicate a decimal point? .font80[(Can be a point or a comma)]
]