Principal speaker
Associate Professor Sama Low-Choy
Many researchers find themselves in the situation where they need to estimate a proportion, for instance of people, events, organizations, or locations, which have something special about them. Some examples of proportions 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 the chance that a rare or pest species may survive in various locations. Traditional statistics courses often present novices with simple ways of analysing proportions, which can be calculated "on the back of an envelope", e.g. binomial test of proportions, or chi-squared tests. Yet inferences from NHT are often mis-interpreted. Furthermore, some journals no longer accept NHT, and 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. To further complicate matters, novices are typically only taught asymptotic approximations to Normality which only apply to large samples.
In this seminar we walk you through NHT and model-based approaches to estimating proportions. We use an interactive role-play designed to engage multiple senses, to help move through the abstract ideas.
Format: Role-play - science fiction adventure including a physical experiment, involving interaction in a large group, and discussion in small groups
Prerequisite: Basic arithmetic: addition, subtraction, multiplication, fractions
Connections: Excellent foundation concepts for any statistical or data science course
Intended audience: All levels welcome.
Researchers from other universities will be joining us for this session, through our collaboration with QCIF (Queensland Cyber Infrastructure Foundation).
This session is for current Griffith University staff and students only.
Event categories
RSVP
RSVP on or before Wednesday 20 November 2024 10.12 am, by email red@griffith.edu.au , or via https://events.griffith.edu.au/x0zNq2