Expanding your inference: transitioning from null hypothesis testing to statistical modelling

Year: 2019

Author: Low-Choy, Samantha, Rose, Judy, Vasco, Daniela

Type of paper: Abstract refereed


Are you wanting to explore the ""new"" statistics? Have you heard of the recent bans on p-values and statistical significance, and want to learn what else you could do?

You may have previously been exposed only to traditional statistics courses, which often present novices with just a single approach of null hypothesis significance testing (NHST) using a binomial test of proportions, or chi-squared tests. Inferences from NHST are often mis-interpreted, which is the main reason that some journals have banned their use. Even model-based approaches that provide confidence intervals present computational difficulties; novices are typically only taught methods, which apply only to large samples.

Here we take the rare approach to place NHST within the context of multiple statistical paradigms: no statistical method is applicable in all situations, and in particular NHST may only be suitable for confirmatory analysis where there is a clear null hypothesis. In addition NHST and p-values can be highly sensitive to sample size, with misinterpretations arising, both for small and large sample sizes. This has led to generations of researchers who are unaware of alternative statistical approaches that may be best suited to a particular applied problem. Inference from a proportion provides a useful and simple context for illustrating the options.


In this workshop we walk you through the use, and practical ramifications, of using 2 different statistical paradigms to approach a simple problem, involving inference about a probability. We use an interactive exercise designed to engage multiple senses, to help beginners to work through the rather abstract notions involved.


Join our Star Trek team to collect and analyse data obtained from ""probes"" sent to a new planet to quantify the %land. This provides a concrete basis for exploring concepts. The two sessions are interactive and will involve discussion in small groups.


We will access the freely available JASP package to do some simple calculations. Please bring your own laptop, fully charged, as this class will involve computing.