Welcome

Welcome to Keiji's Wiki pages! This wiki is (for now) mainly used to:

Manage research projects that I'm involved
Keep track of ongoing analysis
Provide a convenient way to share analytic results with collaborators and clients

In the future, I may be using this wiki to keep my notes on various statistical topics!

Current research projects

Restricted access: Only authorized users can view contents in each project. Projects are sorted in alphabetical order:

But I'm so lazy and I just post links here:

Data dredging (Wikipedia)
- "This is done by performing many statistical tests on the data and only paying attention to those that come back with significant results…"
- Also recommend: Data Torturing by Mills (NEJM)
  - "If you torture your data long enough, they will tell you whatever you want to hear."
Remember: p-values Are Not Effect Sizes (Win-Vector Blog)
When differences in significance aren’t significant differences (Statistics Done Wrong)
- Or maybe read the original paper by Gelman & Stern, 2006 (link to PDF).
The Abuse of Power: The Pervasive Fallacy of Power Calculations for Data Analysis by Hoenig & Heisey, 2001 (American Statistician)
- PDF is available at the author's website
Two sample-size practices that I don't recommend (PDF) by Lenth (2000) criticizes:
- Use of "standardized effect sizes" such as Cohen's d
- Post-hoc power analysis based on the observed effect sizes
- I can't agree more!
Dynamite plots must die (Simply Statistics)
- Let's kill them all.
Don’t go 3D (Fundamentals of Data Visualization)
- "[3D plots] are also almost always inappropriately used." Again, let's kill them all.