3 ms·
"So it’s 99.5% that it occurred due to warming over the industrial era" There are at least a couple of statistical fallacies in this conclusion. And there isn'
by Almaviva 9y ago
"So it’s 99.5% that it occurred due to warming over the industrial era"
There are at least a couple of statistical fallacies in this conclusion. And there isn't a lack of agreement about that.
One problem with p-value tests is precisely that people misunderstand what the p-value means, which is where basic education comes in. It could save people from believing a lot of things they shouldn't. (Like many health and fitness crazes over the last generation, for instance) Or at the very least, we could train science journalists.
- acover 9y agoDo you agree to the statement "so it's 99.5% that we correctly reject the hypothesis it occurred without warming over the industrial era"?
- mnarayan01 9y agoI think you're mainly correct. The main issue is that there's an implied "given that the model from http://www.nature.com/ngeo/journal/v10/n2/full/ngeo2863.html http://www.nature.com/ngeo/journal/v10/n2/full/ngeo2863.html is 'perfect' (at least for this case)". Since it seems unlikely that the model is perfect, the numbers they give are almost certainly inflated.
- acover 9y agoNope. I'm totally wrong.