Biometrics by the Harbour, Hobart, Tasmania, December 1-5, 2015


Pre-conference short courses


These courses have been organised to take place on Sunday November 29, 2015 (9:00am to 5:00pm) at the CSIRO's marine laboratories. The marine labs are situated on Castray Esplanade (just past Salamanca) -- see


Full details of each course are described in the links below.


Your registration fee will include morning and afternoon teas, lunch, and any relevant handouts. Final date for registration is 16 November, 2015. Early bird fees will also apply as for the conference. Fees for full-time students attending the courses will be considerably reduced.

Short course 1: Rosemary Bailey, Chris Brien and Alison Smith - Multiphase experiments: from design to analysis

It is intended that participants will gain the knowledge and skills necessary to design multiphase experiments, evaluate the properties of the designs and analyse the results from multiphase experiments. To this end the course will consist of a combination of presentations and practical sessions. Also, the range of applications in which multiphase experiments occur will be demonstrated.

Further details about this short course can be found HERE and the slides used are HERE.

Short course 2: Adrian Bowman - An introduction to flexible regression for environmental data

Environmental data are often characterised by spatial, temporal and seasonal patterns which are smooth but non-linear in shape. There is a wide variety of approaches to building models which are sufficiently flexible to capture these patterns and some of these will be explored in the course. The emphasis will be on regression approaches which allow the inclusion of additional covariates. Topics will include standard methods of constructing curve and surface representations and the use of additive models, but will also include simple Gaussian process and Bayesian approaches. The level will be introductory, suitable for postgraduate students new to the topics, and the style will emphasise conceptual and modelling issues. A variety of datasets will be considered, with strong emphasis on practical work in R (with which it is assumed participants have some familiarity).


1 Methods of flexible regression

2 Additive models

3 Spatial and spatiotemporal models

4 Case studies

Adrian is supported by the SSAI-WA branch's Frank Hansford-Miller fellowship during his visit.

Short course 3: Richard Emsley - Causal inference in randomised trials

Randomised trials provide a gold standard design for assessing the effectiveness of an intervention or treatment, based on an intention to treat analysis. However, in the presence of departures from random allocation, this suffices to only answer a narrow question about the effectiveness of offering the intervention, based on comparing the average outcome between randomised groups. A series of different questions ask "what is the effect of actually receiving the intervention?", "how does the intervention work?", and "who does the treatment work best for?". These questions require different analysis approaches in order to answer them. This workshop will introduce participants to the statistical methods underpinning the concepts of causal inference in randomised trials. It will include methods for adjusting for non-compliance in randomised trials, using instrumental variables, g-estimation and structural nested models. It will cover recent advances in mediation analysis, and apply all of these methods to randomised trials of complex interventions in mental health. The course level will be introductory and is suitable for postgraduate students and above new to these topics. We will focus on practical examples, modelling issues and the key assumptions, and how these methods can be implemented in standard statistical software, primarily STATA.

Further details about this short course can be found HERE.

Short course 4: Otso Ovaskainen - Analyzing community data with Hierarchical Modelling of Species Communities (HMSC) with R or Matlab

The course provides a tutorial to hierarchical modelling of species communities with generalized linear mixed models, and it is based on applying models to both simulated and real datasets. We use models to address questions about species niches (the relationship between species occurrence and environmental characteristics) and co-occurrence patterns among the species. Further, we examine how species traits and phylogenetic relationships influence species niches and co-occurrence. The basic data involve a matrix on species presence-absences (or abundances) on a set of sites, and some environmental characteristics of those sites. Optional data involve species traits and phylogenetic relationships.

Further details about this short course can be found HERE.

Last updated 21 December, 2015.