Simulating a Study with a Binary Outcome

Introduction I was on Twitter the other day and saw that Andrew Althouse had a nice thread on simulating an RCT and I wanted to expand on his suggestions and provide some of my own recommendations. In Andrew’s thread he explicitly used base R and only default stats functions that are native to R. While these are great tools, they are limited and most researchers wanting to simulate a study will find these tools lacking when trying to simulate their own particular study.

Increasing testing severity in multi-group designs

Introduction A little over a year ago Patrick Forscher wrote a very nice blog post about a simple way to increase the severity of a hypothesis test in a multi-group experimental design (in this case one factor, 3-group design). He highlights the main problem as the following: Multi-group designs are the workhorse of scientific psychology. Multi-group designs apply to any grouping of people, within-person states, situations, or stimuli; interest typically centers around the either the means or conditional means within each group.

Magnitude Based Inference in R and SAS

Updated on: 2020-05-05 Introduction There have been a number of criticisms of “magnitude-based inferences” (Batterham and Hopkins 2006) which is a unique approach to statistics in the sport and exercise science community. As an author of the mbir package (Peterson and Caldwell 2019), I have been watching this all develop closely. What is clear from the criticisms is that MBI has some fatal flaws directly related to the sample size estimations and the interpretations of the probabilities that the MBI spreadsheets provide (Lohse et al.

Welcome to my blog

This is my personal blog area on my website. This is a place for me to express my personal thoughts in a relaxed nature. Most of my writing has been published in academic journals, and most of it is difficult to read and well boring. My hope is this blog will provide me an outlet to communicate directly and informally than my academic writing. What to expect Well, this is the boring part.