I wrote this for the Berkeley Science Review: False positives by Joseph Williams
It's about a recent paper by a faculty in the Business School that presents evidence for how easy it is for studies to report significant results that are really just the result of random noise – false positives.
The image the BSR editors added was excellent, it exactly captures the desperate joy of scientific discovery. There are days I have been shocked at the heights my mood soared to, simply because one of the numbers emitted by SPSS Statistics (the p-value) was lower than 0.05.
I also remember many of the days where the number was greater than 0.05. Those aren't happy memories. More the character building kind.