Did you ever think that Bayes Theorem could apply to politics? Well it does. That is the point of this blog. (If you are not familiar with Bayes Theorem, research it on the internet.)
Bayes was a preacher who wanted to prove that God existed when he started working on his Theorem. None of the statistician of his day accepted his concepts, but later all did. Here is a typical application. We know that only about 5%of people who come to a doctor claiming that they have Lyme's Disease really have the disease. The other complication is that the test for Lyme's Disease i not perfect. The test has a 3% false positive and a 10% false negative.
Here is how Bayes Theorem would apply. We gave two groups being tested. 95% don't have the disease and 5% do have the disease. When we test those who do not have the disease, our test shows falsely that 2.85% have the disease. Of those who have the disease, our test shows that 0.5% wrongly do not have the disease. Now what do we make of all of this?
We have 2.85% of the people being treated for the disease when there is nothing wrong with them and we have 0,5% who should be treated who are not treated. This is what the real world is like in health care. All depends on the rate of occurrence of the disease and the quality of test. Change either and the results will be different. The key to Bayes Theorem is that we need to take into account what we know about the population. This extra knowledge can have a significant impact on the out put of your diagnosis or decision. Treating two different populations as equal can led to a bad diagnosis or decision. This is where politics can get into trouble.
There is a tremendous pressure to treat all groups of people and individuals equally when we all know that there are not. Like in the example, our decision making is not perfect.
PS---I know that this blog is not very clear. I will try better in the future. I hope you get the idea.
Bayes was a preacher who wanted to prove that God existed when he started working on his Theorem. None of the statistician of his day accepted his concepts, but later all did. Here is a typical application. We know that only about 5%of people who come to a doctor claiming that they have Lyme's Disease really have the disease. The other complication is that the test for Lyme's Disease i not perfect. The test has a 3% false positive and a 10% false negative.
Here is how Bayes Theorem would apply. We gave two groups being tested. 95% don't have the disease and 5% do have the disease. When we test those who do not have the disease, our test shows falsely that 2.85% have the disease. Of those who have the disease, our test shows that 0.5% wrongly do not have the disease. Now what do we make of all of this?
We have 2.85% of the people being treated for the disease when there is nothing wrong with them and we have 0,5% who should be treated who are not treated. This is what the real world is like in health care. All depends on the rate of occurrence of the disease and the quality of test. Change either and the results will be different. The key to Bayes Theorem is that we need to take into account what we know about the population. This extra knowledge can have a significant impact on the out put of your diagnosis or decision. Treating two different populations as equal can led to a bad diagnosis or decision. This is where politics can get into trouble.
There is a tremendous pressure to treat all groups of people and individuals equally when we all know that there are not. Like in the example, our decision making is not perfect.
PS---I know that this blog is not very clear. I will try better in the future. I hope you get the idea.
