Bayes Rule Applet
These applets demonstrates Bayes Rule and probability updating in two contexts: a medical diagnosis and employee incentive pay.
The Horrible Disease
How worried should you be if you test positive for some disease?
What does it mean if a test for some disease is "95% accurate"?
Does it mean that, if you test positive, you have a
95% chance of having the disease. While this sounds sensible, the answer is usually "no."
The actual probability depends not only on the reliability of the test, but also
the number of infections in the population to begin with. This applet demonstrates this idea.
Rewarding Employees
It is often difficult to observe effort on the part of employees, so companies are forced to reward employees based on success or failure (a measure of performance) which is only partially controlled by effort. How likely is it that bonuses are going to the bad employees, who simply et lucky, rather than the good ones? This applet answers this question.