Four strategies for dealing with multiple comparisons: Seminar 10 May 2018

Informal, general interest seminars on topics in statistical data analysis

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

Important! Please register here (for catering purposes).