POL90: Exercise 01

class: center, middle, inverse, title-slide

# POL90: Exercise 01
### 
### 2022-01-19

---

## Today's Agenda

.mlarge[

1. Survey: https://forms.gle/AXc92KK4bTs5DNwc9

2. Why literate programming: `R` + `Markdown` + LaTeX +  `knitr`?
  - Merits and challenges

3. Install R, RStudio, LaTeX

4. Install packages

5. Exercise: Play with StatKey, Run your own randomization

]

---

## Literate Programming

.large[
“Let us change our traditional attitude to the construction of programs: Instead of imagining that our main task is to instruct a computer what to do, let us concentrate rather on explaining to human beings what we want a computer to do.” 
— Donald Knuth
]
---

## Why Use `R` + `R Markdown` + LaTeX + `knitr`?

.mlarge[
1. Management 
  - braids together writing, code, output (e.g., tables, plots) in a single document 
    
2. Reproducibility 
  - Dramatically simplifies recreating results in document, a bedrock of good science 
    
3. Elegance 
  - With a little work, much better looking tables, plots, etc. 
    
4. Up-to-date 
  - Changes to data can be immediately reflected in prose, tables, plots. 
]

---
## What Does `R Markdown` Look Like?

RMarkdown template

---
## Why 'Literate Programming'?

.footnote[Artwork by @allison_horst]

---

## Challenges

.mlarge[

1.  Multilingual 
  - Four different languages in one document: `R`, `R Markdown`, LaTeX, `knitr` (glue connecting everything)
    
2. LaTeX 
  - An advanced formatting language that's standard in science but can be complicated.
    
3. Debugging 
  - More languages means more opportunities for bugs; we can help!
    
4. Control 
  - At first you give up some control over formatting.
]

---

## Have you `knit` a `Markdown` document?

.mlarge[

- If not, follow the *setup installation* instructions at: http://appliedstats.org/setup_instructions.html

- To test that it is working, create a new `R Markdown` document in RStudio, and see if can `knit`

- If you can't complete installation, follow-up with instructors

]

---
## Install course packages:

- Copy and paste the following command to your R Console:      
  
    - source("http://appliedstats.org/POL90_2022_packages.R")
    
  - Can also copy from: http://appliedstats.org/setup_instructions.html

- If prompted to restart `R`, click `No`

- If prompted with 'Do you want to install from sources the package which needs compilation?' type `n` and hit `return`

---
## Exercise: Randomization Distribution with StatKey

+ Go to http://www.lock5stat.com/StatKey
  
    - You can just search for StatKey on Google
    
  + Click on "Test for Difference in Means"
  
  + Click on "Leniency and Smiles" for Pop-up menu

- Select "Mosquitos (Beer vs Water)
    
  + Play with randomizing assignment to the two conditions

---
## StatKey: Mosquitos (Beer vs Water)

---
## StatKey: Mosquitos (Beer vs Water)

---
## StatKey: Mosquitos (Beer vs Water)

---

## Chapter 1: Creativity study

.mlarge[

-  Research questions 
 
  - Do grading systems promote creativity in students? 
    
  - Do ranking systems and incentive awards increase productivity among employees? 
    
  - Do rewards and praise stimulate children to learn? 
]

---

## Chapter 1: Creativity study

.mlarge[

- Teresa Amabile ran an experiment on effects of intrinsic and extrinsic motivation on creativity 
 
  - Subjects with creative writing experience randomly assigned: 24 to "Intrinsic," 23 to "Extrinsic" 
    
  - After questionnaire, subjects asked to write a Haiku about "laughter"  
    
  - Poems submitted to 12 poets, who rated them on 40-point scale of creativity 
  - Score is average of 12 judges
]
---

## Idea behind randomization distribution

.mlarge[

- If treatment had no effect, then observed outcomes are unrelated to whether subject was assigned to treatment or control group
 
 - Under assumption of "null hypothesis" or that treatment had no effect, we could shuffle all treatment and control assignments and recalculate difference-in-means

- These simulated results allow us to see if our observed result is extreme compared to other plausible treatment and control groups
 
 - Each randomization is like a possible parallel universe (under assumption of no effect of treatment)
 
 - Questions?

] 
---

## Coding exercise: load packages & data

```r
# load packages
suppressMessages(library(dplyr))
library(janitor)
```

```
## 
## Attaching package: 'janitor'
```

```
## The following objects are masked from 'package:stats':
## 
##     chisq.test, fisher.test
```

```r
library(infer)

# load data
creativity <- Sleuth2::case0101 %>% # chapter 01, case 01
  clean_names()

# view data
creativity %>% head(2)
```

```
##   score treatment
## 1   5.0 Extrinsic
## 2   5.4 Extrinsic
```

---

## Coding exercise: statistical tests

.mlarge[

1. With Base R `t.test()`, calculate whether there is a significant difference-in-means

2. By hand, randomize treatment assignment and calculate a new difference-in-means. See [http://appliedstats.org/chapter1.html](http://appliedstats.org/chapter1.html)

3. Now, repeat Step 2 ten times. How do these results compare to observed difference-in-means?

4. Randomize treatment assignment and calculate difference-in-means 1,000 times, plot histogram of differences

5. With `infer` calculate randomization distribution and plot histogram of differences
]

---

## Have you customized `R` and `RStudio`? (Optional)

.mlarge[

- See R Hacks: http://appliedstats.org/r_keyboard_shortcuts.html

]

---

## Why pipes? (Optional)

---
## What is `clean_names()` and `snake_case`?
  
<img src="images/coding_cases.png" width="100%" style="display: block; margin: auto;" />

.footnote[Artwork by @allison_horst]