Selection Bias

Selection bias occurs when the participants in a study are not representative of the population you want to understand.

It distorts reality. It makes the specific look general.

The WWII Plane Survivor Bias

During World War II, the military looked at planes returning from battle. They saw bullet holes in the wings and tail. They decided to reinforce those areas.

Abraham Wald, a statistician, stopped them.

He said: "You are looking at the planes that came back. You are missing the planes that didn't."

The planes with holes in the engines crashed. They were not in the sample. This is survivorship bias, a form of selection bias.

The "Healthy User" Effect

People who volunteer for health studies are different from people who don't.

They are often healthier. They exercise more. They smoke less. They are more educated.

If you test a vitamin on this group, it might look effective. But maybe these people are just healthy anyway. You cannot apply these results to the general population, who might be sick, sedentary, and poor.

Online Polls

Internet polls are notoriously biased. They only represent people who:

Use that specific website.
Care enough to vote.
Have internet access.

They do not represent "the people." They represent a loud minority.

The Diagnosis

Always ask: "Who is in this study? And more importantly, who is NOT?"

If the sample is flawed, the math does not matter. The conclusion is already wrong.