Workshops in Quantitative Methods -- 2021/22
This seminar series aims to show case a variety of quantitative approaches from a diverse set of disciplines. It hopes to foster discussions about quantitative methods.
We invite you to join us:
Program
27 October 2021 -- Ryan Bakker, University of Essex (Government)
24 November 2021-- Claire Gormley, University College Dublin (Mathematics and Statistics)
26 January 2022 -- Daniel Quintana, University of Oslo (Biological Psychiatry)
6 April 2022-- Christian Arnold, Cardiff University (Politics)
27 April 2022 -- David Lazer, Northeastern University (Political Science and Computer Sciences)
Q-Step Celebrations
Celebration Lecture 2018 - Prof Todd Landman
Q-Step Training the Trainers Programme 2022
Timetable
| Date | Course Leader | Course Name |
|---|---|---|
| Wednesday 16 March | Dr. Marnie Low | Flexible Regression in R |
| Wednesday 23 March | Dr. Nema Dean | Cluster Analysis in R |
| Wednesday 30 March | Dr. Jo Ferrie | Digital Qualitative Methods |
| Wednesday 6 April | Dr. Michael Heaney | Introduction to Social Network Analysis |
| Wednesday 13 April | Dr. Craig Alexander | An Introduction to Data Visualisation in R |
| Wednesday 27 April | Dr. Thees Spreckelsen | Answering questions with Count Data: Introduction to models and R |
An Introduction to Data Visualisation in R
Dr. Craig Alexander
School of Mathematics and Statistics
In this short course, we will introduce elementary graphical methods used for data visualisation. This module will focus on tools within R which allow users to create powerful graphics to aid exploration of datasets. This will begin with the basic plot function but will progress to more advanced graphics using the ggplot2 package. We will cover a variety of example datasets and data structures, including spatial data.
Please note this course will be delivered online via Zoom.
The course will be structured as follows:
A short introductory session and summary lecture material, live on Zoom at 1pm.
- A set of exercises which participants can work through to explore the module. Participants are expected to work through these between both live sessions, in advance of the final discussion forum. Solutions will be shared after the course.
- An online forum, live on Zoom at 4pm, where participants can ask questions on the material with a short review.
For the session, we will use the statistical computing environment R and user interface RStudio. Participants should install both software’s before the session. The software and download instructions can be found at:
R - https://cran.r-project.org/
Rstudio - https://www.rstudio.com/products/rstudio/download/
Please find Eventbrite link here
Time – 1pm-2pm, 4pm-5pm
Answering questions with count data: Introduction to models and R
Dr. Thees Spreckelsen
School of Social and Political Science
How many times is Joe Biden mention in relation to democracy? How many counts of theft are related changing numbers of police officers? These questions are the focus of count data models. This brief introduction will give an overview how to model counts, potential issues involved and how to implement the most common models in R.
Participants should be familiar with R.
Please find link to Eventbrite here
Time: 1pm -3pm
Flexible Regression in R
Dr. Marnie Low
School of Mathematics and Statistics
In this session I will introduce flexible regression methods. The local nature of these methods allows us to effectively model complex non-linear relationships that are often present in data from a variety of backgrounds including environmental and health data. I will introduce and compare different nonparametric regression approaches for handling these complex relationships and show how we can combine multiple non-linear covariates into one model. Often our response variable is not normally distributed and thus I will also cover modelling data with a non-normal response e.g. a binary “survived” response.
Please find here the link to the Eventbrite.
Time: 1pm-4pm
Cluster Analysis in R
Dr. Nema Dean
School of Mathematics and Statistics
Cluster analysis is a set of exploratory unsupervised learning methods that seek to discover a priori unknown group structure in quantitative data. This short course will introduce the participants to the basic idea of cluster analysis and a detailed tutorial on 2 classical and widely-used methods of cluster analysis: hierarchical agglomerative clustering and k-means. Although there will be some technical details, equations will be kept to a minimum and the focus will be building an intuition about the methods. Examples will be demonstrated in R and there will be short lab sessions for the course members to try out the methods in R themselves. The goal is for participants to end the course with an understanding of clustering, an awareness of when it is appropriate to use and when not, an ability to run and interpret a cluster analysis in R, issues and caveats to looks out for, and links to further methods of interest.
Please note that a working knowledge of the R statistical software language (at least the ability to open and manipulate data files, run summaries and plot data) will be assumed. This will not be a course introducing the basics of R.
Please find link to Eventbrite here
Time: 1pm – 4pm
An Introduction to Digital Data and Qualitative Analysis
Dr. Jo Ferrie
School of Social and Political Science
This training is aimed at researchers with familiarity with qualitative analysis (minimum: understanding how to interview, run focus groups etc; thematic analysis) and who wish to learn more about using digital data and analysis approaches. The training will cover two broad themes. The first is around how to conduct traditional qualitative approaches, online. This incudes how to run interviews and adhere to ethics in an online environment. The second theme will involve practical activities and will explore how we can extract data from online platforms and will consider then, the analytical implications. This is an introductory course and the aim is to give some flavour and work with some resources to help students and colleagues get started.
Please find link to Eventbrite here
Time: 1pm – 4pm
Introduction to Social Network Analysis
Dr. Michael Heaney
School of Social and Political Science
This course examines the theoretical and statistical analysis of social networks from an interdisciplinary perspective. Participants learn about the nature, structures, and dynamics of social networks that are relevant to fields such as management, public health, sociology, politics, psychology, economics, and anthropology. The course covers the ontology of networks, major theoretical approaches, common research designs, descriptive statistics, and a variety of techniques for statistical inference. The workshop, conducted virtually over Zoom (link to follow shortly) includes didactic sessions and hands-on computer exercises. The principal goal of workshop is for participants to learn to conduct their own empirical research on social networks on a variety of topics.
Please find link to Eventbrite here
Time: 1pm-4pm