Did you check through the referred papers at the end? It doesn't seem, via reading them, that they are just making shit up. You seem to be taking issue with a turn of phrase in the article, rather than the papers the article refers to. Are you making shit up and didn't actually check if they were making shit up?

Getting between 1% and 84% with differently phrased questions with different people doesn't mean that we can mash the results together and get some number in between.

There being an acceptable way to do so doesn't mean it tells you anything meaningful about the world.

Getting widely divergent numbers means to me that we don't have any accurate understanding.

>Getting between 1% and 84% with differently phrased questions with different people doesn't mean that we can mash the results together and get some number in between.

First stating the range within the existing literature is useful and common practice, at least within academic article writing.

It also doesn't generally mean that any numbers mentioned in the article thereafter are derived by the author mashing the results together and getting some number in between, they are picked from the papers that are referenced at the bottom of the piece.

So here, the wider range is being pulled from one paper that is a wide historical review, the narrower range and estimates for amount of people who hear voices without a recognized mental illness, are pulled from another paper, also a review of existing literature, but one that narrows the scope to more recent research. And then the numbers in the article for the amount of people who link their voices to their God, is the subject of another paper.

In the prior race between almost nobody and almost everybody voted democrat. Not too useful.

If you cannot work out by yourself why an academic article would include that the historical range of the subject under discussion is all over the place, then I am not sure that any attempt of leading you to it is going to help. Perhaps just have a bit more of a think about it. Have you read any of the referenced papers?

If the range is that large either some or all of the studies were poorly done or too different to compare.

It seems that much of what passes for social sciences isn't really science.

So, you do understand why they included the wider historical review then. At least you have got exactly the point they were trying to make by including it in the first place. The literature here historically has been shit and all over the place. Which is the kind of point you might want to make if you are putting forward some other papers based on more recent studies that both opens out the field and narrows the error bars somewhat.