Quan Li. Using R for data analysis in social sciences: a research project-oriented approach

Oxford: Oxford University Press, 2019. — 364 p.Statistical analysis is common in the social sciences, and among the more popular programs is R. This text provides a foundation for undergraduate and graduate students in the social sciences on how to use R to manage, visualise, and analyze data. The focus is on how to address substantive questions with data analysis and replicate published findings. The work adopts a minimalist approach and covers only the most important functions and skills in R to conduct reproducible research. It emphasizes the practical needs of students using R by showing how to import, inspect, and manage data, understand the logic of statistical inference, visualise data and findings via histograms, boxplots, scatterplots, and diagnostic plots, and analyze data using one-sample t-test, difference-of-means test, covariance, correlation, ordinary least squares regression, and model assumption diagnostics.1. Learn about R and Write First Toy Programs
When to use R in a research project
Essentials about R
How to start a prject folder and write our fitst R program
Create, describe, and graph a vector: a simple toy example
Simple real-world example: data from Iversen and Soskice (2006)
Chapter1: R program code
Troubleshoot and get help
Important reference information: symbols, operators, and functionsMiscellaneous Q&As for ambitious readers
Excercises
2. Get Data Ready: Import, Inspect, and Prepare DataChapter 2 program codeMiscellaneous Q&As for ambitious readers
Excercises
3. One-Sample and Difference-of-Means Tests
Conceptual preparation
Data preparation
What is the average economic growth rate in the world economy?
Did the world economy grow more quickly in 1990 than in 1960?
Chapter 3 program codeMiscellaneous Q&As for ambitious readers
Excercises
4. Covariance and Correlation
Data and software preparations
Visualize the relationship between trade and growth using
Scatter plot
Are trade openness and economic growth correlated?
Does the correlation between trade and growth change over time?
Chapter 4 program codeMiscellaneous Q&As for ambitious readers
Excercises
5. Regression Analysis
Conceptual preparation: how to understand regression analysis
Data preparation
Visualize and inspect data
How to estimate and interpret ols model coefficients
How to estimate standard error of coefficient
How to make an inference about the population parameter of interest
How to interpret overall model fit
How to present statistical results
Chapter 5 program codeMiscellaneous Q&As for ambitious readers
Excercises
6. Regression Diagnostics and Sensitivity Analysis
Why are ols assumptions and diagnostics important?
Data preparation
Linearity and model specification
Perfect and high multicollinearity
Constant error variance
Independence of error term observations
Influential observations
Normality test
Report findings
Chapter 6 program codeMiscellaneous Q&As for ambitious readers
Excercises
7. Replication of Findings in Published Analyses
What explains the geographic spread of militarized interstate disputes?
Replication and diagnostics of Braithwaite (2006)
Does religiosity influence individual attitudes toward innovation?
Replication of bénabou et al. (2015)
Chapter 7 program code8. Appendix: A Brief Introduction to Analyzing Categorical
Data and Finding More Data
Objective
Getting data ready
Do men and women differ in self-reported happiness?
Do believers in God and non-believers differ in self-reported
Happiness?
Sources of self-reported happiness: logistic regression
Where to fund more data
References and Readings
Index 3