Replies

There is more possible error here than selection bias. Survivorship bias means there has to be survivors to report incidents. The information has to both exist and is available to include in a study.

Those who fail at using pistols to defend themselves from bear attacks become lunch. People killed by bears become statistics, not study samples.

You may be on to something.

Based on your suggestion Im going to start doing a bit of oppositional to Dean based on the number of self defense reports filed by bears.

There has to be a government grant in there somewhere.

Those who fail at using pistols to defend themselves from bear attacks become lunch. People killed by bears become statistics, not study samples.

Possible, but highly unlikely. People killed by bears are big news. People who are in wild areas tend to look out for each other.

Maybe, from 1870 to 1930... we did not have the printed records, and electric communication systems, before then. Even so, the idea of a person being killed by a bear was a powerful way to start a myth, and people investigated such situations.

About 45-50 people in the Lewis and Clark expedition killed at least 33 grizzly bears over a two year period. Even using relatively low powered muzzleloading rifles and muskets, not one person was mauled or killed, though there were several close calls. Several grizzly bears were shot and not recovered. They are not included in the 33.

On the other side, if a person is successful at stopping a bear attack, and is not injured, there are powerful incentives not to report it (to who?) and not to make a record of it (why?).

I am confident there are far more successful defenses which are not reported/recorded than there are unsuccessful defenses which are not reported/recorded.