Training

This short course will be led by Maria Pampaka

Date: Wednesday, 13 May
Session 1	9.30am – 12pm
Break	12 – 1pm
Session 2	1 – 4pm
Room: Uni Place_4.209

Abstract 

This course introduces participants to measurement theory and the Rasch model for the construction and validation of measures used in social research (and beyond). It provides an overview of the foundations of measurement from an Item Response Theory (IRT) perspective, with particular emphasis on the assumptions and applications of Rasch models.

The Rasch model provides a framework for constructing reliable measures (or scales) from sets of items in tests or questionnaires. Participants will learn how Rasch models can be used to transform raw responses into meaningful measures and how these measures can be evaluated within a validation framework.

The principles governing the application of Rasch models will be illustrated through examples from educational measurement and other disciplines, with clear relevance to research in the social and health sciences. Participants will also have the opportunity to work with different models from the Rasch family, including the Dichotomous Model, the Rating Scale Model, and the Partial Credit Model, and to consider applications to their own research.

The course will be of interest to researchers and practitioners working in areas such as educational assessment, evaluation and satisfaction measurement, tests of skills and knowledge, attitudinal scales, and measures of dispositions and other latent constructs.

The course will combine lectures and practical sessions organised around the following themes:

• Introduction to Rasch measurement;
• the Dichotomous Rasch Model (with software application);
• the basics of validation using the Rasch measurement framework;
• applications of the Rating Scale Model (for Likert-type items) and the Partial Credit Model (for items with different response structures);
• differential Item Functioning (DIF), optimal functioning of rating scales, and dimensionality.

Requirements

Participants should have a basic understanding of introductory statistics. Some familiarity with statistical software and syntax-based commands will be helpful. Example datasets will be provided, but participants are also encouraged to bring their own datasets and measurement problems, possibly with a view to developing a validation report or publication.

This short course will be led by Yan Wang.

Date: Wednesday, 20 May
Session 1	9.30am – 12pm
Break	12 - 1pm
Session 2	1 - 3.30pm
Room: Ellen Wilkison_ A.3.6

Abstract

This course provides an introduction to the quantitative analysis of text from a social science perspective, with a broad range of applications in economics, sociology, communication, and political science. It adopts an applied approach: while theoretical aspects are addressed, the primary focus is to equip students and researchers with fundamental knowledge and practical skills for analysing textual data using basic machine learning methods.

The course helps participants formulate research questions that can be investigated through text data and understand the basic methodologies required to answer them.

Prerequisites and software requirements

Required knowledge 

• Proficiency in the R software environment;
• familiarity with basic statistical concepts.

Recommended background

• Basic understanding of linear algebra;
• basic knowledge of probability theory.

Software installation 

• R
• RStudio

Course structure

The workshop is divided into two main parts:

Lectures (first half)

• Overview of the field and its applications in social sciences;
• fundamental principles of treating text as data;
• basic analytical strategies and their underlying rationales.

Practical sessions (second half)

• Hands-on practice using RStudio;
• textual analysis tasks including:
  • Dictionary-based analysis;
  • classification methods;
  • clustering techniques.

This short course will be led by Diego Perez.

Date: Wednesday, 3 June
Introduction and session aims	10 – 10.30am	Mansfield Cooper 1.02
Theoretical – Bayes	10.30am – 12pm	Mansfield Cooper 1.02
Break	12 – 12.20pm
Theoretical – Bayes Part 2	12.20 - 12.40pm	Mansfield Cooper 1.02
Summary of morning session	12.40 - 1pm	Mansfield Cooper 1.02
Lunch break	1 - 2pm
Practical	2 - 4.45pm	Mansfield Cooper 2.01
Review and closing	4.45 - 5pm	Mansfield Cooper 2.01

Abstract

This course provides a comprehensive introduction to Bayesian statistics, covering the fundamental concepts, programming with STAN, and practical applications in the social sciences. Participants will gain a solid understanding of Bayesian modeling and inference, as well as the skills to apply these techniques to their own research questions.

Prerequisites and software requirements

Required knowledge

• Working knowledge of the R software environment;
• familiarity with (generalised) linear models.

It is recommended that participants have a basic understanding of:

• Linear algebra;
• probability theory.

Software installation 

• R (including tidyverse and lavaan libraries)
• RStudio

In the second half of the workshop, we will be using the STAN platform for Bayesian modelling, accessed through its R interface, RStan. Setting up RStan can be somewhat time-consuming, as it requires the installation of a C++ compiler.

Participants are encouraged to have these software packages installed and configured prior to the start of the workshop to ensure a smooth learning experience. Detailed instructions for the software setup will be provided to registered participants.

This short course will be led by Natalie Shlomo.

Date: Wednesday, 10 June
Session 1	9 – 10.30am
Break 1	10.30 – 11am
Session 2	11:00 – 12:30
Lunch Break	12.30 – 1.30pm
Session 3	1.30 – 3pm
Room: Ellen Wilkinson A3.7

Abstract 

The aim is to provide theory and practical applications of deterministic and probabilistic approaches to data (record) linkage. The course provides a review of best practices and covers theory with respect to the pre-processing requirements, the methods used to calculate matching weights, types of errors in the classification, evaluation procedures and introduction to the E-M algorithm and analysis of linked data.

By the end of the course, students should have an understanding of data linkage techniques and be able to implement and evaluate data linkage procedures. 

Prerequisites 

No pre-requisite is required but students should have an understanding of basic statistical concepts.

References

• Belin, T.R. and Rubin, D. B. (1995) A Method for Calibrating False-Match Rates in Record Linkage. Journal of the American Statistical Association, 90, 694-707.
• Clark D. (2004)  Practical introduction to record linkage for injury research. Injury Prevention, 10(3):186.
• Fellegi, I. P. and Sunter, A. B. (1969) A Theory for Record Linkage, Journal of the American Statistical Association, 64, 1183-1210.
• Gill, L. (2001) Methods for Automatic Record Matching and Linkage and their use in National Statistics, The National Statistics Methodology Series, ONS.
• Herzog, T. N., Scheuren, F. J. and Winkler, W. E. (2007) Data Quality and Record Linkage Techniques. New York: Springer. ISBN 978-0-387-69502-0.
• Lahiri, P. and Larsen, M.D. (2005) Regression Analysis with Linked Data. Journal of the American Statistical Association, Vol. 100, No. 469, 222-230.
• Mason, C.A. amd Shihfen, T. (2008) Data Linkage Using Probabilistic Decision Rules: A Primer, Birth Defects Research (Part A): Clinical and Molecular Teratology 82, 812-821.
• Scheuren, F., and Winkler, W.E. (1993) Regression analysis of data files that are computer matched. Survey Methodology, 19, 39-58.
• Scheuren, F., and Winkler, W.E. (1997) Regression analysis of data files that are computer matched II. Survey Methodology, 23, 157-165.
• Shlomo, N. (2019). Overview of Data Linkage Methods for Policy Design and Evaluation in N. Crato, P. Paruolo (eds.), Data-Driven Policy Impact Evaluation. Springer.
• Winglee, M., Valliant, R. and Scheuren, F. (2005) A Case Study in Record Linkage. Survey Methodology, Vol. 31, Number 1, 3-12.
• Winkler, W. E. (1995) Matching and Record Linkage, in B.G. Cox et al. (ed) Business Survey Methods, New York: J. Wiley, 355-384.

This short course will be led by Tanja Kecojevic.

Date: Wednesday, 24 June
Session 1	9.30am – 12pm
Break	12 – 1pm
Session 2	1 – 4pm
Room: Ellen Wilkinson_ A.3.6

Abstract 

Many research questions in the social sciences focus not only on average outcomes but also on how relationships differ across the distribution of an outcome. For example, the effects of education on wages, socio-economic status on health, or housing characteristics on prices may vary substantially between lower and higher parts of the outcome distribution. Standard regression methods such as Ordinary Least Squares (OLS) focus on conditional means and may therefore obscure important forms of heterogeneity.

This short course introduces quantile regression, a statistical method that allows researchers to estimate and interpret relationships across different points of an outcome distribution. The course will combine conceptual explanations with practical implementation using R. Participants will learn when quantile regression is appropriate, how it differs from standard regression approaches, and how to interpret distributional effects in applied research.

The course will be divided into two parts. The morning session will introduce the conceptual foundations of quantile regression, including the motivation for modelling conditional quantiles, interpretation of coefficients across quantiles, and comparisons with standard regression models. The afternoon session will consist of guided hands-on exercises in R, where participants will estimate quantile regression models using real-world data, compare results to OLS models, and visualise how effects vary across the outcome distribution.

Learning objectives

By the end of the course participants will be able to: 

• Understand the difference between modelling conditional means and conditional quantiles.
• Identify research questions where quantile regression is appropriate.
• Estimate quantile regression models using R.
• Interpret coefficients across different quantiles of the outcome distribution.
• Visualise and interpret heterogeneous effects across the outcome distribution.

Prerequisites

Participants should have:

• Basic familiarity with regression analysis, such as OLS.
• Some experience using R for statistical analysis.
• Rand RStudio installed on their laptop prior to the workshop. No prior experience with quantile regression is required.  using R with real-world datasets.

Additional information

The course is designed for researchers and postgraduate students in the social sciences who are interested in modelling heterogeneous or distributional effects in their data. Quantile regression is widely used in fields such as economics, sociology, health research, and education, particularly in studies of inequality and outcomes that exhibit substantial variation across their distribution.

All teaching materials, example datasets, and R scripts will be provided to participants in advance to facilitate learning and reproducibility. The course is designed to be reusable and could be delivered in subsequent years with updated datasets or applied examples.

Cancellation by the University

The University of Manchester reserves the right to cancel a short course ten (10) days before if there are not sufficient delegates registered on the short course or any associated event at its sole discretion.

In the event of such a cancellation the University will refund the value of the booking or any ticket sold upon proof of purchase.

The University expressly excludes any liability for any direct or indirect losses or damages howsoever arising as a result of such cancellation and will not, for example, be responsible for any travel or accommodation costs incurred.

In the event of cancellation, the University will use reasonable endeavours to publicise the cancellation and details will be posted on the website associated with the short course.

Attendees are responsible for checking this information prior to the event.

Wherever possible, the University will endeavour to contact delegates by email in the event of cancellation.

Cancellation by you

We will refund the value of the booking or any ticket sold upon proof of purchase when cancellation takes place at least three weeks before the short course.

When cancelled less than three weeks but no more than seven (7) working days before the short course we will refund a percentage not less than 50% of the charge.

No refunds will be given for cancellation less than seven (7) working days before the short course date or for non-attendance without notification, unless a refund is made entirely at the discretion of the University.

Notice of cancellation must be sent to cmi@manchester.ac.uk.