This
site is under construction. Please check back every few weeks for
updates
COMMON MISTEAKS
MISTAKES IN
USING STATISTICS: Spotting and Avoiding Them
Introduction
Types
of Mistakes
Suggestions
Resources
Table
of Contents
About
Mistakes Involving the Limitations of Frequentist Inference
Procedures (Hypothesis Tests and
Confidence Intervals)
The reasoning involved in frequentist hypothesis tests and
confidence intervals is not simple. Sometimes people are taught "just
the gist," which may be adequate up to a certain point. However, such
oversimplification often leads to misinterpretations and
to assuming that studies establish more certainty
than they really do.
So, in order to understand the mistakes discussed in this section, it
may be helpful first to read:
then
then
You may need to go
back and forth between these three pages; having three screens open or
printing out the three pages might help. Rest assured that most people
need to go through these ideas at least twice to really understand
them, so don't give up if you don't catch on instantly. Then go
on the to following pages, looking
back at the previous three as needed to clarify any misunderstandings
that may crop up.
Misinterpretations of p-values
Type I and Type II Errors
Mistakes Involving Power
The File Drawer Problem (Publication Bias)
Multiple Inference
Data Snooping