LEL Statistical Training Workshops (STeW)

The Linguistics and English Language Statistical Training Workshops (STeW) is a series of workshops that runs throughout the academic year and is aimed at students and staff who would like to advance their statistical knowledge and skills.

On this page you can find information on the current plan for the a.y. 2024-25. Attending requires you register and registration links are sent out via email at the beginning of each semester. Further workshops might be added later during the year.

Most workshops are recorded and can be watched on the LEL Quantitative Methods channel on the UoE Media Hopper Create.

If you have any question, you can get in touch with Dr Stefano Coretta.

1 2024-25

1.1 Semester 1

Quantitative Methods in LEL (QML)
Date Full teaching term
Time See timetable
Level 🟢 Beginners
Description This is the flagship stats course of the Linguistics and English Language department. It covers the basics of statistics in R and PhD Students are particularly encouraged to audit it.
Prerequisites None
Materials https://uoelel.github.io/qml/

1.2 Semester 2

Ordinal models for Likert/rating-scale data
Date Thursday 23 January 2025
Time 15.00–17.00
Level 🔴 Advanced
Description In this workshop you will learn how to model data from Likert/rating scales, such as the ones commonly used in developmental and psycho- linguistics using ordinal models.
Prerequisites Need to be familiar with regression/linear models in R including models with random effects and binomial/Bernoulli (aka logistic) regressions. While a background in Bayesian inference is useful, the workshop will explain the basics of Bayesian inference in the context of ordinal models.
Materials TBA
Prospective power analyses for frequentist regression models
Date Thursday 27 March 2025
Time 15.00–17.00
Level 🟠 Intermediate
Description You will learn how to perform prospective power analyses for regression/linear models fitted with the frequentist lme4 package. Power analyses are a necessary (albeit always neglected) component of frequentist analyses, since they help the researcher determine the required sample size. Much of the replicability crisis we are facing is due to lack of statistical power (aka low sample sizes) in virtually all published studies.
Prerequisites Need to be familiar with regression/linear models in R including models with random effects and binomial/Bernoulli (aka logistic) regressions.
Materials TBA
Mindless statistics and the “null ritual”
Date Thursday 1 May 2025
Time 15.00–17.00
Level 🟢🟠🔴 Any level
Description In this workshop we will go through bad statistical practices that are common even among experienced researchers and how to avoid them. In particular, we will talk about the “null ritual”, which is a invalid approach to frequentist statistics commonly used in research. Most of the content of the workshop will be drawn from Gigerenzer’s Mindless Statistics paper.
Prerequisites Some familiarity with frequentist statistics is useful but not necessary. This workshop is thought for both beginners and experienced researchers or students, whether they normally use quantitative, qualitative or mixed methods.
Materials TBA

2 Previously

intRo: Data visualisation with R
Date 18 December 2022
Level 🟢 Beginners
Description This is an introduction to R and data visualisation for absolute beginners. If you want to learn R for the first time, this is a good start.
Prerequisites None.
Materials https://intro-rstats.github.io
Recording Day 1, Day 2
Introduction to Bayesian Linear Models
Date Thursday 26 October 2023
Time 15.00–17.00
Level 🟠 Intermediate
Materials https://stefanocoretta.github.io/learnBayes/
Recording Watch
Including random effects in Bayesian Linear models with brm()
Date Thursday 16 May 2024
Time 15.00–17.00
Level 🟠 Intermediate
Description This workshop is targeted especially to MSc students who have attended the QML course.
Prerequisites Anybody who has attended the previous Bayesian workshops or who has experience with basic linear models with brm() is welcome to join.
Materials https://uoelel.github.io/brm-group/
Recording Watch
Generalised Additive Mixed Models (GAMMs)
Date Thursday 18 April 2024
Time 15.00–17.00
Level 🟠 Intermediate
Description GAMMs are useful to analyse historical data, time-series data or data that could have non-linear effects (like language development data).
Prerequisites It assumes some familiarity with linear models fitted with lmer()/glmer() or brm().
Materials https://stefanocoretta.github.io/learnGAM/
Recording Watch (recording of previous run of the workshop)
Bayesian Priors in Linear Models
Date Thursday 22 February 2024
Time 15.00–17.00
Level 🔴 Advanced
Prerequisites Need to be familiar with Bayesian Linear Models including Random Effects and/or have attended the Introduction to Bayesian Linear Models workshop. This is not suitable for students who have only attended the QML course (see below for dedicated workshop).
Materials https://stefanocoretta.github.io/learnBayes/ (Session 2)
Recording

Unfortunately I forgot to record this, but you can watch the following videos from another workshop to learn about priors (these videos go into much more detail!).