What Statistics 101 Doesn't Teach, But Should

Charles J. Geyer

School of Statistics
University of Minnesota

Introductory statistics courses teach the math not the philosophy,
partly because the math is hard enough and partly in order not to
offend, because statistically questionable methodology is widely
used in science.  This questionable methodology was highlighted
in recent articles titled "The truth wears off: Is there something
wrong with the scientific method?" in the New Yorker and
"Why most published research findings are false" in PLoS Medicine.
Both parts of the New Yorker title are wrongheaded.
The apparent decline of "statistical significance" in follow-up studies is
a well understood result of publication bias and multiple testing without
correction, and it does not show anything wrong with "the scientific method."
The PLoS Medicine title perhaps needs an addendum: in areas of science
that rely heavily on statistics (which is understood in that journal).
We will explain these issues and discuss the problems they raise for
scientific inference that are primarily statistical
and discuss the electric power lines and cancer debacle as an example.