Your Guide to P Values and Confidence Intervals – Shoe Size, Penile Length, Football and Heart Attacks!

Posted on January 7, 2017

In work and life, we are always attempting to make links with the hundreds and thousands of variables around us. Some of these links may be true while others may just be false alarms. How do we know which links are significant enough for us to change our behaviour?  This is the whole basis of statistics.

It is important to recognise that the only certainty in life is the uncertainty and complexity that surrounds us. In order to make sense of this uncertainty we make certain assumptions and adjust as best as we can for uncertainty.
I’ll take the example of medical studies. In most studies, we want to know if an intervention delivers a positive result or if a particular exposure causes a disease. The first thing we do when we want to investigate two variables is to state a hypothesis. This hypothesis is usually framed as a null hypothesis. So let’s take an example of a fun study.

References

Shoe Size and Penile Length
Shah, J., & Christopher, N. (2002). Can shoe size predict penile length?. BJU international, 90(6), 586-587.
Football and Heart Attacks
Wilbert-Lampen, U., Leistner, D., Greven, S., Pohl, T., Sper, S., Völker, C., … & Steinbeck, G. (2008). Cardiovascular events during World Cup soccer. New England Journal of Medicine, 358(5), 475-483.
ASA statement

http://www.amstat.org/asa/files/pdfs/P-ValueStatement.pdf.