I have statistics question. I have an (astronomical) image, and I want to sum up the pixels within a square, and also get the error of this sum. I use the program DS9 (in case somebody knows it), and it does give me a sum, and an error. But the strange thing is, the error is exactly the square root of the sum. And I have no idea why this is the case. I've never such an error computation.
Does anybody know more?