Monthly Seminars

Informal, general interest seminars on topics in statistical data analysis

​​

2018 - August

Regression to the mean and how to deal with it

Despite being “transparent to the point of being obvious”, the phenomenon of regression to the mean – first recognised in the 1800s – is still catching out researchers. We’ll look at what regression to the mean is, how to identify it, how you can reduce it at the design stage and statistical methods you can use to account for it at the analysis stage of your pre-post experiment.

 
The talk will be about 30 minutes long and will be followed by (free!) afternoon tea and time for discussion.  Please register by Thursday, 9th August, 10.00 am, so we can cater properly for afternoon tea!
 
Speaker:  Mark Donoghoe, Stats Central, UNSW
 
Date and Time:  Thursday, 9th August 2018, 3:00 pm to 4:00 pm
 
Location:  Central Lecture Block, Theatre 3
 
Important! Please register here (for catering purposes).
 
 
 

2018 - July

Analysing small data sets - a mini overview

Analysing data sets with a small number of observations can be a challenge! This seminar will touch on some of the problems that arise in these analyses and outline a number of analysis methods, together with their limitations. The general types of method we will talk about are parametric (e.g. t test, ANOVA), non-parametric (e.g. Mann-Whitney, Kruskal-Wallis) and randomisation methods.

The talk will be about 30 minutes long and will be followed by (free!) afternoon tea and time for discussion.  Please register by Thursday, 12th July, 10.00 am, so we can cater properly for afternoon tea!

Date and Time:  Thursday, 12th July 2018, 3:00 pm to 4:00 pm
 
Location:  Civil Engineering Building (H20), Room G1
 
Important! Please register here (for catering purposes).
 

2018 - June

Intraclass Correlations - How to use them for reliability and why there are so many of them

Before a measure can be used for research or application purposes, the measure itself must be shown to be replicable across measurements.  Intraclass correlations (ICC) are one way to assess reliability.  This seminar will discuss the calculation of ICCs, how to report the results and show why there are so many different versions.

The talk will be about 30 minutes long and will be followed by (free!) afternoon tea and time for discussion.  Please register by Thursday, 14th June, 10.00 am, so we can cater properly for afternoon tea!
 
Speaker: Nancy Briggs, Stats Central, UNSW
 
Date and Time:  Thursday, 14th June, 2018 - 3.00 pm to 4.00 pm
 
Location:  Central Lecture Block, Theatre 3
 
 
 
 
 
 

2018 - May

Four strategies for dealing with multiple comparisons

Multiple hypotheses may be generated by

  • multiple treatment arms;
  • heterogeneous treatment effects;
  • measuring multiple outcome variables.

In a hypothesis testing framework, using p < 0.05 as a criterion for declaring significance, it can be easy to get results that are significant by chance when many hypotheses are tested. This talk will discuss four things you can do when faced with multiple comparisons and will cover:

  1. the difference between controlling the family-wise error rate and the false discovery rate;
  2. the Bonferroni-Holm adjustment;
  3. the Benjamini-Hochberg adjustment;
  4. strategies for multiple outcome variables and strategies for multiple comparisons which are correlated.

The talk will be about 30 minutes long and will be followed by (free!) afternoon tea and time for informal discussion. Please register by Thursday 10th May, 10.00 am, so we can cater properly for afternoon tea!

Speaker: Eve Slavich, Stats Central, UNSW

Date and Time:  Thursday, 10 May 2018 - 3.00 pm to 4.00 pm

Location:  Central Lecture Block, Theatre 4
 
Slides for the presentation are here.