-
Please do.
I read something earlier in all of this. It was great. But if someone else could explain it simply, to try and help out. That would be grand.
Chalfie-
How simply?
On the most basic level, it's like this:
Say you get a positive result. This could be due to you being positive and the test working correctly (true positive), or you being negative and the test working incorrectly (false positive).
Say you get a negative result. This could be due to you being negative and the test working correctly (true negative), or you being positive and the test working incorrectly (false negative).
In terms of being one sample out of an entire population, you have a certain probability of being positive or negative (a priori probability). The test then also has a certain probability each of producing the four different scenarios outlined above (true / false positive, true / false negative).
So when you test people and get positive / negative numbers, what you get is the combination of these different probabilities, and you don't just know which bit is due to what.
SwissChap
It's worth noting that due to how Bayesian probability works, having a test that is 60% accurate does not mean that, should you test positive, you have a 60% chance of having had Covid19.
Unless you know the true prevalence of infection in the population you cannot state with any accuracy what your chance of having had it are.
The maths is counter intuitive and I certainly cannot write out an explanation here. Happy to Google links for people...