Wednesday, January 16, 2013

Why You Shouldn't Believe 100% of What You Read

One of my favorite subjects in college and something that is both abhorred and misunderstood by 99.999% of the population. Rest assured, when anyone uses statistics, you should feel very confident in throwing the BS flag for the following simple reasons.

1. 79% of statistics are made up on the spot (or is it 78%)
2. If they aren't making it up, the person quoting some statistic has a 90% chance of not knowing where the statistic came from.
3. If they know where it came from, there is a 95% chance that they haven't actually read the source material.
4. If they have read the source material, there is a 37% chance it was made up (see rule 1), or if the source is Facebook, blog, or comment forum, this number jumps to 93%.
5. If the source material is accurate, the quoter still has a 82% chance of not understanding what the statistic was actually measuring.
6. If they understand what the statistic was actually measuring, there is a 61% chance that the situation they are applying the statistic to was never intended because of the exceptions or assumptions within the study.
7. If the statistic is applicable, there is less than a 2% chance that the quoter has logically thought through the implications of the statistic and realized that there is a 66% chance that A) it is irrelevant or B) it actually means the opposite of what was thought.
8. More than likely, the statistic contradicts some other strongly held belief of the quoter and he will simple ignore it in that situation.

So in summation to the quoter's favor there is a (.21*.9*.05*.07*.18*.61*.02*.66)= 9*10^-5 % chance of them being right, hence the need to throw the BS flag more often. This is the same as flipping heads on a coin 20 times in a row. Sure it happened in Godot, but this is real life. If you could be assured of those odds, it would make playing the lottery worth it.

No comments:

Post a Comment