Overview
This page gathers all the important information about the course content and schedule. Each week is dedicated to a specific set of topics and you are expected to complete readings from the course textbook and other homework of a particular week by the Tuesday lecture of that week (so complete readings and homework of Week 2 by the Tuesday lecture of Week 2, and so on). There will be no workshops in Week 1.
Each week there will be a 50-minute lecture and a 2-hour workshop. The lecture expands on the most important and/or difficult topics of the week, but you are expected to learn and understand all the topics of the week as treated in the readings from the course textbook. Before each lecture, students are expected to work through the assigned textbook chapters in full. This includes not only reading the text, but also completing all accompanying R/RStudio activities, code examples, installation steps, exercises, and quizzes as you encounter them. In the workshop you will further practice your R skills with group data challenges.
The Schedule overview below gives you an overview of each week, while the Weekly schedule is a more in-depth description of the activities and materials of each week.
Schedule overview
PDF version of the slides
The lecture slides linked below are HTML documents that are rendered in your browser. If you wish to download a PDF version of them, click on the hamburger menu (bottom-left corner), click on the cog “Tools” and select “PDF export mode”. Then, use the method for your specific browser to print the page to PDF.
The following table is a schedule overview of the course, with links to lecture slides and workshop instructions. Due dates for assessments are also included. See Assessments for more info.
Topics overview
In the following table, you can find a more detailed overview of the topics for each week, divided into research methods, statistics and R specific topics.
| 1 |
|
|
|
| 2 |
Research methods |
Quantitative data analysis |
RStudio, R basics, R packages and reading data |
| 3 |
Inference |
Data summaries |
R scripts and summarise data |
| 4 |
Research cycle |
Data viz principles |
Quarto documents, data transformation and plotting |
| 5 |
Bayesian inference |
Probabilities and Gaussian models |
Gaussian models with brms |
| 6 |
|
Regression modelling (numeric predictor) and posterior draws |
Regression modelling with brms, MCMC draws |
| 7 |
Null Ritual |
Regression (categorical predictor) |
Treatment coding and contrasts, posterior draws of the expected values |
| 8 |
|
Bernoulli and log-normal regression |
Bernoulli and log-normal regression, model diagnostics |
| 9 |
Open research |
Regression (multiple predictors) and interactions |
Categorical/categorical and categorical/numeric interactions |
| 10 |
What’s next? |
Outlook on further statistical topics |
Numeric/numeric interactions |
Course textbook
The course textbook can be found at the following link (also available from the site menu): Quantitative Data Analysis for Linguists in R. The textbook has been written by Stefano, tailored specifically for this course.
Course data
The course data is available for download on the Data for Quantitative Methods in Linguistics website. For each data set, an entry gives detail about the study and the data set. When using data for the first time, you should read the related entry, abstract and skim the linked paper (if applicable) for context.
Week by week
Week 1: Onboarding
Intake form
- You must complete the intake form before coming to the Tuesday lecture.
- The link to the form can be found on the Learn website.
Week 2: Quantitative methods and R basics
Questions
- What are the components of research methods?
- What makes a good research question and research hypothesis?
- What are the three steps of quantitative data analysis?
- What is the computational workflow of quantitative data analysis?
Skills
- Think critically about research methods and research questions.
- Master the basics of RStudio and the programming language R.
- Learn how to install and use R packages.
- Learn how to read a variety of data files in R.
Course website
Carefully read the homepage.
Familiarise yourself with this Course content page, read the Overview and the Schedule overview.
- You must always complete the week’s readings by the Tuesday lecture of that week (this rule applies to all the readings of the course, so it will not be repeated).
- This includes not only reading the text, but also completing all accompanying R/RStudio activities, code examples, installation steps, exercises, and quizzes as you encounter them.
- Read and complete any activity in the Preface and the Week 2 chapters of the course textbook.
Week 3: Inference, uncertainty and data summaries
Questions
- What is inference and why do we need it?
- Why are uncertainty and variability important?
- What is and isn’t statistics?
- What are summary measures?
Skills
- Using R scripts to keep reproducible code.
- Reading tabular data in R.
- Summarise data with
summarise() and count().
- Get grouped summaries with
group_by().
- Read and complete any activity in the Week 3 chapters of the course textbook.
Week 5: Probability distributions and Gaussian models
Questions
- What are probabilities and probability distributions?
- How do we describe and visualise probability distributions?
- How do we use Gaussian probability distributions to estimate a mean and standard deviation?
- What is the Bayesian approach to probability and inference?
Skills
- Produce density plots.
- Use
pnorm() and qnorm() to obtain probabilities and quantiles of Gaussian distributions.
- Fit Bayesian Gaussian models with brms.
- Report results from Bayesian Gaussian models.
- Read and complete any activity in the Week 5 chapters of the course textbook.
Week 6: Regression models
Due Thursday at noon
Complete Quiz 1 this week, to check your learning before moving on onto the second part of the course.
The link to the quiz is available on the Learn site of the course.
Questions
- What are regression models and what are they for?
- How do we interpret intercept and slope in a regression model?
- Why do we need Markov Chain Monte Carlo to estimate models?
- How do we use MCMC draws from a model fit?
Skills
- Fit regression models of the form \(y \sim x\) with brms in R.
- Interpret the regression coefficients table of the model summary.
- Extract and plot MCMC draws.
- Report results from regression models.
- Read and complete any activity in the Week 5 chapters of the course textbook.
Week 7: Categorical predictors and frequentist statistics
Due Thursday at noon
Submit your self-reflection, covering Weeks 4-6.
Questions
- How do regression model work with categorical predictors?
- How are categorical predictors coded with contrasts?
- What are frequentist statistics and the “Null Ritual”.
- Why researchers should prefer Bayesian statistics?
Skills
- Fit regression models with a categorical predictor with brms in R.
- Interpret the regression coefficients table of the model summary.
- Extract and plot MCMC draws of models with categorical predictors with two or more levels.
- Report results from regression models with categorical predictors.
- Read and complete any activity in the Week 7 chapters of the course textbook.
Week 8: Binomial/Bernoulli and log-normal regression models
Questions
- How can we model binary outcomes with regression models?
- How can we model numeric outcome variables that are only positive?
- What are model diagnostics?
- Which model diagnostics should I pay attention to?
Skills
- Fit and interpret regression models with a Bernoulli family in brms.
- Fit and interpret regression models with a log-normal family in brms.
- Use model diagnostics to understand the model’s goodness of fit.
- Apply standard solutions when models do not converge.
- Read and complete any activity in the Week 8 chapters of the course textbook.
Week 9: Open Research and regression model interactions
Questions
- What is Open Research?
- What makes research reproducible, replicable, robust and generalisable?
- What are Registered Reports?
- How can we model interactions between predictors in regression models?
Skills
- Think about Open Research practices and how to assess existing research in light of Open Research practices.
- Distinguish the four ways research can be reliable.
- Use Research Compendia.
- Fit and interpret regression models with categorical/categorical and categorical/numeric interactions in brms.
- Read and complete any activity in the Week 9 chapters of the course textbook.
Week 10: What’s next?
Due Thursday at noon
Submit your self-reflection, covering Weeks 7-9.
Questions
- When continuing my learning journey, which further statistical topics should I first focus on?
Skills
- Fit and interpret regression models with numeric/numeric interactions in brms.
- Read and complete any activity in the Week 10 chapters of the course textbook.
Week 11
Complete Quiz 2 this week, to check your learning across a variety of topics covered in the course.
The link to the quiz is available on the Learn site of the course.