FundRmentals 01

# FundRmentals 01
### Dr Danielle Evans | University of Sussex

---

# Session Overview

#### Part 1: Course Overview

- What even is R & why are you making me learn it?

- Introduction to the fundRmentals course

- The R-cult: why R is ⭐*magical*⭐

#### Part 2: Getting StaRted

- Tips for success

- Next steps

- Installing R & RStudio

---

# Part 1: Course Overview

---

# What even is R anyway? 🤔

- R is a programming language where we can talk to our computer & tell it what to do...

```r
print("Hi LearneRs!")
```

```
## [1] "Hi LearneRs!"
```

]

- RStudio is a more user-friendly interface where we can use R

]

---

# Introduction to fundRmentals

#### Course Structure

- Weekly tutorials to do in your own time

- Weekly 'HappyHouR' practical sessions

<br>

#### Course Aims

- Gain solid understanding of the 'fundRmentals'

+ Feel confident supervising project students using R

+ Feel inspired to use R/RStudio for your own projects

]

#### FundRmentals [Webpage:](https://fundrmentals.netlify.app/)

<!--

-->

]

---

# The R-cult: why R is ⭐*magical*⭐

- Makes complicated things easy to do

- Super efficient

- Collaboration

- Reproducibility

- Visualisations

- Modelling

- Reporting

- Presenting

- & *sooo* many other cool things!!

---

# A *very* simple regression

- 1 line of code, much quicker than clicking in SPSS:

```r
lm(outcome ~ predictor, data = data)
```

<br>

### *"But Dan, what if I have 10 predictors?"*

- Still just 1 line of code...

```r
lm(outcome ~ predictor_1 + predictor_2 + predictor_3 + predictor_4 + predictor_5 + predictor_6 +
     predictor_7 + predictor_8 + predictor_9 + predictor_10, data = data)
```

---

# A *very* simple regression

- 1 line of code, much quicker than clicking in SPSS:

```r
lm(outcome ~ predictor, data = data)
```

<br>

### *"But Dan, what if I want to do a t-test?"*

```r
t.test(outcome ~ grouping_variable, data = data)
```

### *"Or a correlation?"*

```r
cor.test(~ variable_1 + variable_2, data = data)
```

---

# A *very* simple regression

- 1 line of code, much quicker than clicking in SPSS:

```r
lm(outcome ~ predictor, data = data)
```

<br>

### *"But Dan, what if I have a factorial design?"*

- Still just 1 line of code...

```r
lm(outcome ~ predictor_1*predictor_2, data = data)
```

---

# A *very* simple regression

- 1 line of code, much quicker than clicking in SPSS:

```r
lm(outcome ~ predictor, data = data)
```

<br>

### *"But Dan, what if I have a stupidly complex model?"*

---

# A Gross Latent Growth [Model](../img/parmodel.jpg)

## Model

]

.center[
.teeeny[
[Evans & Field, 2020](https://royalsocietypublishing.org/doi/10.1098/rsos.200422)
]
]
]

# Code

]

- Just 5 lines of code for this monstrosity:

```r
model <- "i =~ 1*sat140 + 1*ks2_mat + 1*ks3_mat + 1*ks4_maths
s =~ -4*sat140 + 0*ks2_mat + 3*ks3_mat + 5*ks4_maths

i ~ sex + int_sdq_11 + iq_cent + wm_cent + ses_cent + cse_voc + cse_olevel + cse_alevel + cse_degree + par_mh_fa + mum_home_int + part_home_int + cai_cent + psai_cent +
home_teach_1_al + home_teach_1_nu + schl_supp_sa + schl_supp_se + eas_harmony + eas_control

s ~ sex + int_sdq_11 + iq_cent + wm_cent + ses_cent + cse_voc + cse_olevel + cse_alevel + cse_degree + par_mh_fa + mum_home_int + part_home_int + cai_cent + psai_cent +
home_teach_1_al + home_teach_1_nu + schl_supp_sa + schl_supp_se + eas_harmony + eas_control"

growth(model, data = data, estimator = "MLR")
```

]
]
]

---
# Data Viz: A Pretty, Plain, APA Scatterplot

```r
ggplot(data, aes(x = flipper_length_mm, y = body_mass_g)) +
  geom_point() + 
    labs(x = "Flipper Length (mm)", y = "Body Mass (g)") + 
      theme_apa() 
```
]
]

]

---
# Data Viz: Grouped Scatterplot

```r
ggplot(data, aes(x = flipper_length_mm, y = body_mass_g, colour = island)) +
  geom_point() + 
    labs(x = "Flipper Length (mm)", y = "Body Mass (g)", colour = "Island") + 
      theme_apa() 
```
]
]

]

---

# Data Viz: sCATterplot

```r
ggplot(data, aes(x = flipper_length_mm, y = body_mass_g)) +
 geom_cat(cat = "lil_bub", size = 2) + 
  labs(x = "Catnip Consumed", y = "Happiness") + 
    theme_apa()
```

]]

]

---

# RMarkdown

- Combine code (analyses, plots etc.) & text in one document

- Great for writing notes about your analyses or for creating nicely formatted word docs/htmls/pdfs etc.

- Super useful for writing journal articles, dissertations & theses

- Additional packages (`papaja` & `rticles`) for apa style/journal formatting

- **Most (if not _all_) students will use RMarkdown for their analyses**

]

]

---

# Inline Code

- Allows you to include code within the main text of your document

- Can be used with test results, values from tables etc.

- Any values will be automatically updated when you render the file

- Incredibly useful & efficient

- Helps you avoid tYpos & rounding erroRs

<br>

**The final sample consisted of `` `r nrow(data)` `` participants.**

**The final sample consisted of 500 participants.**

]

---

# You can create beautiful [slides](https://github.com/yihui/xaringan) like these...

![](../img/office.gif)

---

# You can create super cool [websites...](https://rstudio.github.io/distill/) 😉

<!--

-->

---

# Part 2: Getting StaRted

---

# Tips for Success

#### For Supervision...

Number 1 rule: don't panic! We've all got this!!

- Students have access to everything we've covered thus far on Canvas 
- They are equipped to be able to do their analyses - they just don't know it 
- They'll also have access to an R helpdesk next semester & two analyses workshops

#### For Yourself...

- Practice as often as you can
- Be kind to yourself - it feels really slow & confusing when you first start
- Ask questions, often someone else explaining it slightly differently makes everything click
- There's a huge online community of R users, for any problems/errors you're having the solution can usually be found quite easily by searching
- Google is your best friend

---

# Next Steps

#### Complete the [tutorial](http://milton-the-cat.rocks/learnr/r/r_getting_started/) before our first .orange[HappyHouR] next week, which covers:

- Installation & tour of RStudio

- Working in RStudio

- Installing & loading packages

- RMarkdown

- Getting data into RStudio

- Tables in RMarkdown

---

# Installing R & RStudio Walkthrough