Waving Hello to R

R Workshop 2024 in UTokyo

SUN Yufei（Adrian）

UTokyo & Tsinghua University

About Me

PhD Candidate in Political Science, Tsinghua University
Visiting Researcher at ISS
Research interests: Political Psychology, Natural Language Processing
Dissertation: How and why Hong Kong uses CCP discourse, and its empirical impact

Founder of Tsinghua R Workshop, 5-year R Workshop instructor
Github Campus Expert

About the Workshop

Week 1: Basic grammar
Week 2: Data cleaning & Basic visualization
Week 3: Loops and batch processing
Week 4: R & Quarto (for HTML, PDF, Word, etc.)

Outline

R Basics
R Packages
Data Import
Data Export
Data Structure
Variable Extraction
Variable Characteristics
Variable Attributes

What is R? → Just like your 📱

R → Your Phone

R Studio/VS Code → iOS/Android

R packages → Apps

R Basic → iMessage

packages from other source → Line

function → Message Sending

object → Contact

:::

R Studio/VS Code → iOS/Android

IDEs (Integrated Development Environments)

R Studio/VS Code → iOS/Android

and … VS Code/pycharm/Positron

latest plugins
multiple languages

Not customized for R
Unstable

R packages → Apps

install

install.packages("adrianlp")
remotes::install_github("syfyufei/adrianlp")

call

library(adrianlp)
require(adrianlp)
adrianlp::tokenlize(text_data)

Basic R & Packages

R Basic → iMessage

Data Manipulation
- Create, subset, and manipulate data frames
- Use functions like subset(), merge(), apply()
Statistical Analysis
- Perform basic statistical tests
- Use functions like t.test(), lm(), summary()
Data Visualization
- Create basic plots and charts
- Use functions like plot(), hist(), boxplot()
Programming
- Write functions and loops
- Use control structures like if, for, while

Classic R Packages → Line

dplyr
- Data manipulation and transformation
- Functions: filter(), select(), mutate(), summarize()
ggplot2
- Data visualization
- Functions: ggplot(), geom_point(), geom_line()
tidyr
- Data tidying
- Functions: gather(), spread(), separate(), unite()
shiny
- Interactive web applications
- Functions: shinyApp(), fluidPage(), server()

function → Message Sending

Functions in R

Built-in Functions
- Predefined functions available in R
- Examples: mean(), sum(), length()
User-defined Functions
- Custom functions created by users
- Syntax: function_name <- function(arg1, arg2) { ... }

Components of a Function

Function Name
- The identifier used to call the function
- Example: my_function
Arguments
- Inputs to the function, specified within parentheses
- Example: function(arg1, arg2)
Body
- The code block that defines the function’s operations
- Enclosed within curly braces { ... }
Return Value
- The output of the function, specified using return()
- Example: return(result)

function → Message Sending

Code to Access `lm()` Help Documentation

Use help() function
- Code: help(lm)
Use ? symbol
- Code: ?lm

Usage

lm(formula, data, subset, weights, na.action,
   method = "qr", model = TRUE, x = FALSE, y = FALSE, qr = TRUE,
   singular.ok = TRUE, contrasts = NULL, offset, ...)

Let’s Practice (Hot Potato)

Hot Potato(“ばくだんゲーム”; “击鼓传花”) but one by one version

Hot Keyboard Time

Question

How to display the help documentation for the sum() function? Provide two methods.

Answer

help(sum)
?sum

`<-`

Assignment operator, the shorthand for the assign() command

Syntax: <variable name> <- <object>

aValidObject <- 1:5
aValidObject

Why `<-`？

Intuitive
Will not be confused with “=”
Shortcut input
- PC: Alt + -
- Mac: option + -

a <- 12
25 -> b

Naming Rules

Don’t start with a number (Error: 1stday).
No special symbols except . and _ (Error: M&M).
Case sensitive (X != x) ! means "not"/"no",!=` means “not equal to”.
Don’t override built-in commands if necessary(avoid: list <- c(1:5)).

Hot Keyboard Time!

Data Input

Built-in Data

data()

Hot Keyboard Time!

###Data Types That Can Be Read Directly

.RDS (single object)
.RData (multiple objects)
.txt
.csv

Syntax: <name><- <read command>(<data path>)

df_rds <- readRDS("aDataset.rds")
df_txt <- read.table("D:/aDataset.txt")
df_csv <- read.csv("./aDataset.csv")

Data Types Need To Call The Package To Read

Call the package through library or require, and then use the commands in it.

# SPSS, Stata, SAS
library(haven)
df_spss <- read_spss("<FileName>.sav")
df_stata <- read_dta("<FileName>.dta")
df_sas <- read_sas("<FileName>.sas7bdat")  

# Quick Import of Forms
library（reader）
df_csv <- read.csv("<FileName>.csv")
df_table <- read.table("<FileName>.csv/txt")

# Excel
library(readxl)
df_excel <- read_excel("<FileName>.xls")
df_excel2 <- read_excel("<FileName>.xlsx")

# JSON (JavaScript Object Notation)
library(rjson)
df_json <- fromJSON(file = "<FileName>.json" )

# XML/Html
library(xml)
df_xml <- xmlTreeParse("<url>")
df_html <- readHTMLTable(url, which=3)

Hot Keyboard Time!

Data Input

The Swiss Army Knife of data reading:rio

library(rio)
df_anything <- import(<AnyTypeOfData>)

Data Output

Syntax: (, file = )

Saving as R Data

saveRDS(df_toy, file = "df_toy.rds")
save(df_toy, ls_monks, file = "test.rdata")

Saving as CSV File

write.csv(df_toy, file = "toy.csv")

Note: If your data contains CJK characters, you may encounter encoding issues when saving as a CSV file.

Data Munging

load("/Users/adrian/Documents/Yufei_Sun/THU/projects/slides/course/rworkshop_in_UTokyo/wvs7.rda")

Data Structure

Observations
Variables
Data Structure

Data Structure

load("/Users/adrian/Documents/Yufei_Sun/THU/projects/slides/course/rworkshop_in_UTokyo/wvs7.rda")

wvs7

nrow(wvs7) # Get the number of rows in the data
ncol(wvs7) # Get the number of columns in the data
names(wvs7) # Get the variable/column names
str(wvs7) # Get variable names, types, number of rows and columns

Variable Extraction

To extract a column from a data frame, there are at least two common methods:

[a,b]

wvs7[, "country"]

<data frame>$<variable name>

wvs7$country

Variable Characteristics

numeric variables

table

table(wvs7$age)

summary

summary(wvs7$age)

Non-numeric Variables: Summary Tables

table(wvs7$female)
table(wvs7$marital)

For factor variables, we can also extract their level information

levels(wvs7$religious)
levels(wvs7$marital)

Variable Attributes

Attributes are attributes of all types of variables.

length(wvs7$age) # Get the length of the variable (here it is the number of rows)
unique(wvs7$age) # Get the unique values of the variable

summary(wvs7$age) # Get all the above information about the year
class(wvs7$age) # Check the structure of the year: vector, matrix, array, dataframe, list
typeof(wvs7$age) # Check the type of the year elements

Summary

summary(wvs7$age)
summary(wvs7)

Take-Home Points

R Basics
- RStudio/VS Code are like operating systems (IDEs)
- R packages are similar to apps
- Functions in R are comparable to “Message Sending”
Loading Packages
- install.packages(): Installing packages
- library() or require(): Loading packages
- help() or ?: Accessing function documentation
Data Import:
- readRDS(), read.table(), read.csv(): Reading various file formats
Data Export:
- saveRDS(), save(), write.csv(): Saving in different formats
- rio::import(): The “Swiss Army Knife” for data reading

Data Structure:
- nrow(), ncol(): Get number of rows and columns names(): Get variable/column names
- str(): Get variable names, types, number of rows and columns
Data Extraction:
- dataframe[, "variable"] or dataframe$variable: Extracting variables
Variable Characteristics:
- table(), summary(): Examining variable characteristics
- levels(): Understanding factor variables
Variable Attributes:
- length(), unique(), class(), typeof()

Thank you

Github Page: https://github.com/syfyufei

Email: sunyf20@mails.tsinghua.edu.cn

Personal Website:https://syfyufei.github.io/

Waving Hello to R

About Me

About the Workshop

Outline

What is R? → Just like your 📱

What is R? → Just like your 📱

R Studio/VS Code → iOS/Android

R Studio/VS Code → iOS/Android

R packages → Apps

install

call

Basic R & Packages

R Basic → iMessage

Classic R Packages → Line

function → Message Sending

Functions in R

Components of a Function

function → Message Sending

Code to Access lm() Help Documentation

Usage

Let’s Practice (Hot Potato)

Hot Keyboard Time

Question

Answer

<-

Why <-？

Naming Rules

Data Input

Built-in Data

Data Types Need To Call The Package To Read

Data Input

Data Output

Saving as R Data

Saving as CSV File

Data Munging

Data Structure

Data Structure

Variable Extraction

Variable Characteristics

numeric variables

Non-numeric Variables: Summary Tables

Variable Attributes

Summary

Take-Home Points

Thank you

Code to Access `lm()` Help Documentation

`<-`

Why `<-`？