# Step 1 — wait until done
update.packages(ask = FALSE)
# Step 2
install.packages("pacman")
# Step 3
pacman::p_load(
tidyverse, # Data manipulation
ggplot2, # Figures
wbstats, # World Bank data
estimatr, # Regression for weighted data
modelr, # Turn lm() results into a tibble
furniture, # Row means
modelsummary, # Regression and balance tables
ivreg, # IV 2SLS
countrycode, # Recode country names
remotes, devtools)
# Step 4
remotes::install_github("jrnold/masteringmetrics", subdir = "masteringmetrics")
# Step 5
remotes::install_github("avehtari/ROS-Examples", subdir = "rpackage")
# Step 6
remotes::install_github("vdeminstitute/vdemdata")
# Step 7
remotes::install_github("xmarquez/democracyData")
# If a remotes::install_github(...) fails, try from a non-KU internet connection.
# Step 8
pacman::p_load(wbstats, rosdata, masteringmetrics, vdemdata, democracyData)About & setup
This advanced statistics course in sociology, Multiple Regression and Fundamentals of Causal Inference, gives you a working understanding of both the theory and the application of multiple regression analysis. You will learn to interpret regression coefficients and to use regression to test hypotheses and make predictions.
Beyond regression itself, the course builds a solid grasp of the principles of causal inference: telling correlation from causation, spotting confounders, and using research designs — randomized controlled trials, quasi-experiments, and observational studies — to estimate causal effects. You get hands-on experience doing this in R, on real sociological questions such as the effect of news-media consumption on perceived ethno-racial discrimination, or of police interventions on repeat domestic violence.
The steps below must be done before the first session, in order — whether or not you have used R before.
Before the first session
- Get a recent version of R (≥ 4.6.0).
- Install RStudio (≥ 2026.06), or update to the most recent version. See also this guide.
- Open RStudio and set up a Project for this class (we walk through this together in the first session — the setup slides will be linked here once they are ready).
- Run the lines below step by step in the RStudio Console. This can take a while. If asked “Do you want to install from sources the package which needs compilation?” answer No. Wait until the cursor blinks behind a
>again before running the next step.
- Type
sessionInfo()and check that these appear under “other attached packages”: tidyverse, ggplot2, wbstats, estimatr, modelr, furniture, modelsummary, ivreg, countrycode, rosdata, masteringmetrics, vdemdata, democracyData. If any is missing, ask a tutor to help you set up.
Always bring your laptop to class.
Course components and expectations
Methodological and substantive lectures introduce the key concepts and methods; exercise classes give you hands-on practice. Four expectations:
- Prepare each class by completing the assigned readings.
- Engage actively — ask questions, join discussions, complete the exercise-class tasks.
- Online quizzes: solve the weekly Absalon quizzes (at least 10 to qualify for the exam).
- Home assignment (final exam): submit by noon, 14 January 2027.
Course structure
A weekly three-hour lecture every Wednesday at 10:00 — two lecture blocks, small exercises, and a 15-minute break. A tutoring session every Friday offers a smaller setting for stats problems and hands-on R. Each week there is an Absalon online quiz — individual; you must submit at least 10 of the 13 to qualify for the exam.
Schedule & team
| When | What | Where | Who |
|---|---|---|---|
| Wed 10–13 | Lecture | CSS 1-1-18 | Merlin |
| Fri 08–10 | Exercise group | CSS 2-2-49 | Sofie |
| Fri 10–12 | Exercise group | CSS 2-0-30 | Natalie |
| Fri 10–12 | Exercise group | CSS 2-0-42 | Joakim |
Course literature
Two textbooks, both available at Academic Books on City Campus and the university library:
- Stats: Data and Models, Global Edition — De Veaux, Velleman & Bock (2021), Pearson. (Reference for regression and statistical inference, carried over from last semester.)
- Mastering ’Metrics: The Path from Cause to Effect — Angrist & Pischke (2014), Princeton University Press. (How regression identifies causal effects.)