1
Course Notes
1.1
What have I signed up for?
1.1.1
Data Science Can Help Social Scientists
1.1.2
Course Goals
1.2
Setup in R
1.3
First Time Working in R and RStudio
1.3.1
Open RStudio
1.3.2
Using R as a Calculator
1.3.3
Working in an R Script
1.3.4
Preparing your R script
1.3.5
Executing Commands in your R script
1.3.6
Objects
1.4
Assignment 1
1.5
R Markdown
1.5.1
How to get setup in RMarkdown
1.5.2
Additional RMarkdown resources
1.6
Wait a second why R?
2
Description
2.1
Process of Describing
2.1.1
Example Process
2.2
Summarizing univariate data
2.3
Functions to summarize univariate data
2.3.1
Using functions in R (overview)
2.4
Loading data into R
2.4.1
Working with datasets in R
2.4.2
Measuring the Turnout in the US Elections
2.4.3
Getting to know your data
2.5
Comparing VEP and VAP turnout
2.5.1
Creating new variables in R
2.6
Comparing Presidential vs. Midterm turnout
2.6.1
R shortcut for writing vectors
3
Causation with Experiments
3.1
What separates causation from correlation?
3.1.1
Potential Outcomes Framework
3.1.2
Causal Effects
3.1.3
Fundamental Problem of Causal Inference
3.2
Randomized Controlled Trials
3.2.1
Experiments: Why Randomize?
3.2.2
Experiments: How to Analyze
3.2.3
Ingredients of an Experiment
3.3
Application: Is there racial discrimination in the labor market?
3.3.1
Variable classes
3.4
Making tables
3.4.1
Crosstabulation
3.5
Conditional Means
3.6
Relational Operators in R
3.7
Subsetting data in R
3.7.1
Getting Booooooooolean
3.8
Creating New Variables using Conditional statements
3.8.1
ifelse statements
3.9
Types of Experiments
4
Visualization
4.1
Application: Social Status and Economic Views
4.2
Boxplots
4.2.1
Data Summary: Boxplot
4.3
Barplots
4.3.1
Saving Plots
4.3.2
Creating New Variables
4.4
Application: Changing Minds on Gay Marriage
4.4.1
Creating new variable
4.4.2
Using ifelse to create new variable
4.4.3
Calculating the Average Treatment Effect
4.4.4
Visualize means in a barplot
4.5
Scatterplots
4.6
Histograms
4.6.1
Happy research ending
4.7
Application: Trends during COVID
4.8
Line Plots in R
4.9
Causal claims from before vs. after comparisons
4.10
Visual tips and tricks
4.11
Common R plotting functions and arguments
4.12
A note on
ggplot
5
Causality with Non-Experimental Data
5.1
Why can’t we always experiment?
5.1.1
What can we do instead?
5.2
Causal Identification Strategies
5.2.1
Three Common Identification Strategies
5.2.2
Threats to Cross-Section Designs
5.2.3
Threats to Before-After Designs
5.2.4
Threats to Diff-in-Diff Designs
5.3
Application: Economic Effects of Basque Terrorism
5.3.1
Applying 3 Identification Strategies
5.4
Placebo Tests
5.5
Wrapping Up Causality
6
Loops in R
6.1
The anatomy of a loop
6.1.1
The key parts of a loop
6.1.2
A short example
6.1.3
Troubleshooting a loop
6.1.4
Your turn
6.2
Application: U.S. Supreme Court
6.2.1
Troubleshooting the loop
6.2.2
Visualizing the Results
6.2.3
Enhancing the plot
6.2.4
Wrapping Up
7
Prediction
7.1
Prediction Overview
7.2
Process of Prediction
7.3
Example: Forecasting 2020 US Election based on 2016 Results
7.3.1
Choose Approach
7.3.2
Assess Accuracy
7.3.3
Iterate to improve predictions
7.4
Example: Using polls to predict the 2020 election results
7.4.1
Choose an approach: Let’s analyze some polls
7.4.2
Loop through all states
7.4.3
Check Accuracy
8
Prediction with Regression
8.1
Regression in the wild.
8.2
Application: Baseball Predictions
8.3
Step 1: Approach- Regression in R
8.3.1
Visualizing a regression
8.3.2
Making predictions with regression
8.4
Step 2: Checking accuracy of model
8.4.1
Root Mean Squared Error
8.5
Step 3: Iterate and Compare Models
8.5.1
Regression with Multiple Predictors
8.6
Application: Predicting Campaign Donations
8.6.1
Visualizing the results
8.6.2
Step 1: Calculate Predictions
8.6.3
Step 2: Check Accuracy
8.6.4
Step 3: Iterate
8.6.5
Adding Model Predictors
8.7
Uncertainty with Prediction
8.7.1
Example: Butterfly Ballot in Florida
8.7.2
Multiple Predictors
8.7.3
Confidence Intervals
8.8
Cross-Validation
8.8.1
Cross-Validation Process
8.8.2
Application: Forecasting Election Results
9
Fairness and Ethics
9.1
Application: Criminal Justice
9.1.1
Load data
9.1.2
Prediction/Classification process
9.1.3
Step 1: Regression Model
9.1.4
Step 2: Check Accuracy
9.2
Taking Fairness Seriously
9.2.1
Extended Learning
10
Uncertainty
10.1
Hypothesis Testing Overview of Process
10.2
Sampling and Uncertainty
10.2.1
Sampling Distribution
10.3
Z-scores and p-values
10.3.1
Relationship to Confidence Intervals
10.4
Wrapping up the Process
10.5
Application: Health Savings Study
10.6
Additional Applications
10.6.1
Example Using Regression
10.6.2
Example Using prop.test()
10.7
In-Class Exercise Questions
11
Text as Data
11.1
Why text?
11.2
R Packages for text
11.3
Application: State of the Union
11.3.1
Cleaning Text
11.3.2
Preparing a Corpus
11.3.3
Word Frequency
11.4
Word Importance
11.5
Additional Descriptive Statistics
11.5.1
Dictionary Analysis
11.6
Application Programming Interfaces
11.6.1
Twitter API v1
11.6.2
Saving R Objects
11.7
Application: Sentiment Analysis
11.7.1
Searching tweets for keywords
11.7.2
Sentiment Analysis Overview
11.7.3
Implementing sentiment analysis
11.7.4
Your Twitter Analysis
12
Mapping
12.1
Why maps
12.1.1
To map or not to map?
12.2
Mapping in R
12.2.1
Using ggplot2 with maps
12.3
Choropleth Maps
12.4
Application: 2021 NJ Election Results
12.5
Application: Voter Identification Laws
12.5.1
Using the
%in%
function
12.6
Your turn to map
12.7
Application: Terrorist Attacks in France
12.7.1
Adding points to a map
12.8
Animating Data
12.9
Application: Spread of Coronavirus
12.9.1
Mapping Animation with World Map
12.10
Application: Territorial Control of Syria
13
Choose Your Own Adventure
13.1
How to investigate new data
13.1.1
Dealing with Uninformative Variable Names
13.1.2
Dealing with Missing Data
13.1.3
Dealing with Variable Codings that Aren’t Quite Right
13.1.4
Dealing with Parts of Datasets
13.2
Where to Go Next
13.2.1
Network Analysis
13.2.2
Additional Topics to Explore
13.2.3
That’s all
Data Science for Political Science
5.5
Wrapping Up Causality
Do you get this joke?
