Brought to you by QCIF in partnership with Griffith University Researcher Education and Development and delivered by Griffith University senior statistician A/Prof Sama Low-Choy, these four half-day workshops provide insight into a wide variety of statistical concepts and applications.
Places for these one-off workshops are very limited and only available to researchers from QCIF member institutions. Please use this form to apply for places at one or more of these workshops. Applications are now closed – thank you for your interest.
Session 1: Understanding Probability – Putting Sharks on Trial
Tuesday 23rd May 10.30am – 2.30pm. Online workshop.
This interactive tutorial addresses a fundamental topic in statistics, which is subtle and often overlooked: the meaning of probability.
This tutorial focuses on the meaning of probability in statistics, which is important for various statistical approaches used in today’s world. Probability is often introduced in introductory statistics courses in the context of null hypothesis testing, significance testing, and classical statistical modeling. It is also a key concept in Bayesian statistical models and machine learning algorithms for big data analysis. Hence understanding probability can enhance all quantitative research.
Through role-play and dialogues about an environmental dilemma, we will help you understand the meaning of probabilities and become aware of common logical fallacies. This will enable you to avoid misconceptions about the interpretation of statistical analyses and will enhance your understanding of statistical concepts across different areas of research.
Relevant for any quantitive research project, this four hour introductory workshop requires nothing more than basic maths knowledge and a willingness to join in with group activities.
Session 2: Expanding your Statistical Universe: from Null Hypothesis Testing Towards Statistical Modelling
Tuesday 13 June, 10.30am – 2.30pm. Online workshop.
Pre-requisites: “Understanding Probability” workshop or equivalent background
Many researchers find themselves in the situation where they need to estimate a proportion of things – for instance of people, events, organizations, locations – that have something special about them. Examples of such a proportion might include: the percentage of people with an issue or needs regarding health, justice or education; or the proportion of locations suffering from environmental impacts, or where rare or pest species occur.
Traditional statistics courses often introduce just a single approach of null hypothesis significance testing (NHST) that can be calculated “on the back of an envelope”: binomial test of proportions, or chi-squared tests. But inferences from NHST are often mis-interpreted, and some journals no longer even accept NHST and instead demand model-based approaches (e.g. estimates of effect sizes or differences). However, the mathematics and computation of model-based approaches for estimating proportions can be much more challenging than for continuous measurements.
In this seminar we walk you through NHST and model-based approaches to estimating proportions. We use an interactive role-play designed to engage multiple senses, to help move through the rather abstract notions involved.
Session 3: Baby Steps Towards Bayes Theorem
Tuesday 8th August, 10.30am – 12.30pm. Online workshop.
Pre-requisites: “Expanding your Statistical Universe” workshop or equivalent background
We disrupt the traditional approach to teaching Bayesian statistical thinking that relies on the mathematics of integrals. Ferrie’s boardbook for babies (and their parents) presents a much simpler way of demonstrating Bayes’ theorem in action. There are lots of pictures, few words and no integral signs!
We start by looking at how some cookies may or may not have candy on them, which affects whether a bite out of a cookie has candy or not. Then we show you how to “flip” this perspective around, so that if you take a bite, and you know if it has candy or not, you can talk about the chance that the cookie had candy or not. We explore the same idea in other contexts, like interpretation of medical tests (e.g. for viruses) or sabermetrics for Aussie Rules.
Session 4: Cookies-and-Candy Exploration of Bayesian vs. Classical Statistics (Bayes for Toddlers)
Tuesday 22nd August, 10.30am – 12.30pm. Online workshop.
Pre-requisites: “Baby Steps Towards Bayes Theorem” workshop or equivalent background
This is an extension to “Bayesian Probability for Babies”. In his book, astrophysicist Chris Ferrie tells a story about candy on cookies. He considers the chance of a bite with no candy on it, and works up to exploring the chance that the cookie itself has no candy at all. In fact, this is a clever, visual metaphor for differentiating the Frequentist and Bayesian concepts of probability. In this workshop, we will work through the yummy, tangible landscape of cookies and candy (as well as other visual landscapes) to explore the simple arithmetic involved. Along the way we will explore the broader points of difference and similarity between Bayesian and classical statistical thinking, and how these help you in your research.
Places at these workshops are available by application only, using this form. Please note, we expect demand to be high and may not be able to offer places to all applicants. Applications are now closed – thank you for your interest.