One of the questions posed by religious people is how atheists propose to disprove the existence of a God. On the face of it, the question is ridiculous as there is no reason to believe God would be the default position; it is not on the atheist to disprove God before positive evidence is given in favour of a God. From here, the conversation runs the risk of misunderstanding falsification and positive evidence.
Falsification was articulated by Karl Popper. His argument was that good ideas are ones that are prohibitive: if true, the idea draws a clear distinction between what can happen and what cannot. Anything an idea proposes cannot happen is then looked for. The standard example is swans: if the idea is that all swans are white, then the idea proposes no non-white swans can happen. Then, you seek a non-white swan. If you’ve looked hard enough and long enough and not found a non-white swan, but have found many more white swans, then you can provisionally accept that idea as knowledge.
This does not mean that for someone to reject an idea they must know how to falsify the idea and then do it. Falsification is actually a metric by which we see whether an idea is a valid one, even before we set out to disprove it. Falsification encourages us to define an idea well enough to know what we might see, should the idea be false. So, contrary to standard practice, the reasonable question is for atheists to ask religious people how they would disprove God. It is only in an intelligible answer to that question can we even be sure the idea the religious person is peddling is valid.
And this brings us to the opposite of falsification: positive evidence. This is evidence that conforms to the nature proposed by an idea. Given a sufficiently ill-defined idea, like Freudian psychology, everything fits the proposed nature of the idea. This is why the idea must first be defined in such a way as to explicitly prohibit something, as to be (possibly) falsified. Then, this well defined idea must have some positive evidence presented in its favour. An idea cannot be considered knowledge simply on the existence of this positive evidence; proper attempts at falsification must be made as well. But, without the evidence in favour of the idea ‘that which can be asserted without evidence can be dismissed without evidence’.
So, it is religious people who must answer the question ‘how would you disprove a God’ to demonstrate their idea is an intellectually valid one, and then they must present positive evidence for that God. This may seem unfair, but if the religious person is presented a coherent case with a well-defined idea, it should be the easiest thing in the world.