statistics
6 posts · all topics
-
Why studying hospital patients can make two unrelated diseases look linked
In 1946 the statistician Joseph Berkson showed that if two conditions each make hospital admission more likely, a sample of inpatients can show a link between them that does not exist outside the hospital. The same trap, now called collider bias, turns up wherever a sample is chosen by the very things being compared.
-
How cancer survival can improve in every stage while no patient lives longer
In 1985 a team led by A. R. Feinstein showed that better scans had made lung cancer survival look better within every stage, without changing how any patient fared. The effect, named after a Will Rogers joke, later turned up in prostate cancer grading too.
-
Why making a model bigger can make it worse, then better again
Classical statistics says test error should rise once a model is flexible enough to memorise its training data. Experiments since 2018 show a second fall in error beyond that point, called double descent; the pattern is well documented, but how much it matters in practice is still argued over.
-
Why guessing a batter's average gets better when you look at other players
In 1956 Charles Stein proved that when you estimate three or more averages at once, using each one's own data alone is never the best you can do on total error. Pulling every estimate toward the group mean does better, a result statisticians still call Stein's paradox.
-
Was the hot hand a fallacy after all?
For thirty years the hot hand in basketball was a textbook example of people seeing patterns in randomness. In 2018 two economists showed that the standard test was biased against finding streaks, and that correcting it reversed the classic result.
-
How reliable are AI text detectors?
Tools that claim to spot machine-written text are used to judge essays and applications. The research suggests they misfire most on some human writers, can be evaded by rewording, and that even a small error rate produces many false accusations when most writing is human.