BLOG

I write about R, Python, and statistics.

What is it like to run a Registered Replication?

Registered replications are generally considered the gold standard of replication studies. What does it take to actually run one?

We introduce some major improvements.

Using the constraints of Likert-scale data to fully reconstruct it.

There’s a point at which more information yields negative returns.

How to set up custom qualification tests to screen MTurk workers.

How to get set up with AWS and Mechanical Turk.

Getting your materials ready for public release.

How do you know where to start?

The best way to know if something works is to test it.

Combine executable code, math formatting, and nice markdown in one easy place.

Violin plots are a way to present more information about your raw data.

Bootstrapping is a powerful and intuitive technique.

How much variability can you expect from the effect size estimate in a single study?

A look inside an open science workflow.

Solving a probability question with simulation.

Take your effect sizes on the go.