Course information

• Instructor: Ryan Carlin (rcarlin@gsu.edu)

• Teaching Assistant: Ozlem Tuncel (otuncelgurlek1@gsu.edu)

• Meeting Time: 4:30-6:00 pm, Tuesday

• Class Location: Langdale Hall 1081

Ozlem’s Office Hours and TA Sessions

• My office: Langdale Hall 1027
• Office hours: 2:30-4:00 pm every Monday and Thursday

Slides, Notes, and Tips

Week 1: Syllabus Overview

✔️ Goal: Review of the syllabus and last semester.

Class materials

Week 1 Slides

Ozlem’s notes from Week 1 class

Software and others

✔️ Goal: Make sure you are familiar with basics of R.

⚠️ Our library offers online R workshops, and I highly recommend them!

Review of R

I recommend Adam Kuczynski’s (University of Washington) to review R ▶️ Adam’s Guide to R

Learning LaTeX

I encourage all of you to get familiar with LaTeX or similar kind of document preparation system to typset your problem sets. GSU offers online/in-person LaTeX course. I use Overleaf for typetting these sort of documents. Recently, I have been using Quarto in R and Phyton to typeset reports and presentations. Here are some useful links to learn LaTeX:

• GSU Library’s LaTeX workshop
• Visit Dr. Fix’s website for his LaTeX presentations.
• Overleaf’s 30-minute guide
• Overleaf’s detailed 3-part guide

You can alternatively learn and use R Markdown or Quarto. Here are some useful links:

• Using Quarto in R
• Posit’s (previously known as R Studio) guide to R Markdown

Week 2: OLS Review

Class materials

Week 2 Slides

Ozlem’s notes from Week 2 class

R scripts for Week 2: Script I & Script II

More info on distributions and OLS

• Here is a website that shows how to use plots to identify OLS assumptions.
• Here is a comparison of probability mass and density functions.

Week 3: MLE Introduction

Class materials

Week 3 Slides and Carlin’s Notes

Ozlem’s notes from Week 3 class

We do not have any R script for this week.

Week 4: Binary Logit

Class materials

Week 4 Slides

Ozlem’s notes from Week 4 class

R scripts for Week 4: Script I & Script II

Important Note

❗You can find the change in the syllabus on iCollege.

Software and others

• This is a great guide on logit and probit although the software is Stata.
• This is another great source to see how to perform logit and probit in R, Stata, SAS, and SPSS.
• I recommend checking Zelig in detail if you are going to use logistic regression a lot.
• I also recommend this Shiny app for comparing logit, rare events, and Firth penalized MLE models.

Week 5: Workflow in PoliSci

Class materials

Week 5 Slides

Notes from Workflow lecture

• I know I bombarded you with a lot of information on GitHub, LaTeX, and R Studio. Do not be discouraged if you encounter any issues when using these programs. Make sure to read the error message, take a deep breath, and let Google help you.
• If you want to practice the GitHub exercise by yourself, check the links on my slides, or watch this YouTube video that I based my own presentation.
• Overleaf only allows GitHub and Dropbox version control for premium accounts. Yet, if you are using LaTeX software in your PC, you cannot connect your projects to GitHub like the R project I showed.
• See resources part in my website for CV, journal article, and Beamer templates in LaTeX. Check Academic and Software for templates. Hint: search latex CV template or beamer template in GitHub, you will find a lot of templates that you can fork, copy-paste, and use freely. If I need inspiration that’s what I do most of the time.

Software Notes

• I learned using GitHub with R using this Happy Git with R book.
• I created these two bug fix files for Zelig and DAMisc packages that are not on CRAN anymore. Zelig lazy load & DAMisc lazy load

Week 6: Interactions

Class materials

Week 6 Slides

Ozlem’s notes from Week 6 class

R script for Week 6: Script I

Software and others

• Interpreting log-odds is tricky! Hence, I have these sources for you to help with interpretation. Source 1, Source 2, Source 3, Source 4, Source 5, Source 6, Source 7, Source 8
• Also, I recommend reading articles that use marginal effects and predicted probabilities for inspiration. Here are 4 different examples that I found for you: Brambor, Clark, and Golder, 2017, Kavasoglu 2021, Green and Haber 2006, Kluver and Spoon, 2016. These might not be about your own research, but you might find them useful.

Week 7: Ordinal Models

Class materials

Week 7 Slides

R script for Week 7: Script I

Software and others

• Interpretation of ordinal models is tricky! So, I have a couple of resources you might want to check. Source 1, Source 2, Source 3, Source 4, and Source 5.
• Training Computational Social Science PhD Students for Academic and Non-Academic Careers
• Here is a nice application of ordinal models: Williams et al. 2021

Week 8: Multinomial Models

Class materials

Week 8 Slides

R script for Week 8: Script I & Script II

Software and others

• I have a couple of resources you might want to check for interpretation. Source 1 (dfidx package), Source 2 (nnet package), Source 3 (mlogit package), Source 4, Source 5, Source 6.

Week 9: Survival Analysis

Class materials

Week 9 Slides

R script for Week 9: Script I

Important changes on deadlines

• Week 11 (October 31) – Ozlem R session for binary logit
• Week 15 (November 28th Tuesday to December 1 Friday) – Final exam
• Week 15 (December 5) – Final papers due

Software and others

• Most comprehensive list of packages and functions related to hazard models: Source 1
• If you are planning to use survival analysis in your final paper, read this book by Box-Steffensmeier & Jones. This little green book is also great source.
• Emily Zabor’s guide on survival analyis is super comprehensive and helpful.
• Great explanation of left and right cencoring.
• Several other sources that I found really helpful for understanding survival analysis theoretically and application in R: Source 1, Source 2, Source 3, Source 4, Source 5, Source 6, Source 7
• Example article for survival analysis

Week 10: Count Models

Class materials

Week 10 Slides

R script for Week 9: Script I & Dataset

Software and others

• Overdispersion explained with an example: here
• Better explanation of GLM with zero inflation: here
• Null vs residual deviance meaning: here
• Several resources useful for conducting count analysis in R: Source 1, Source 2, Source 3, Source 4, Source 5, Source 6, Source 7
• Two article examples: Example 1 and Example 2

Week 11

Class materials

Week 11 Slides

Software and others

• We do not have any R script to run for this week, but I will be doing a session where I replicate one of the problem sets.
• Two things can be useful this week. The first one is the grammar of graphics. You really need to understanding underlying layers of the ggplot so that you can confidently modify your figures.
• The second one is the EDA. I have few other sources that might help you to be knowledgable about your data: source 2 and source 3.
• Here is a stackoverflow answer to understand the different between fixed-effects and clustering: read Alex P. Miller’s answer here
• Another great source for fixed effects vs clustering is this Abadie et al. (2017) paper which tells the same thing with the previous answer: article link
• This book by Huntington-Klein is super useful if you want more info on fixed-effects: book link

Week 13: Student Presentations

Sources

• If you need templates for LaTeX, I have different templates prepared in my website.
• Here is a great article on how to present better: Smith and Salmond, 2011.