There are always going to be exceptions to the rule, usually due to a strong counfounding factor that can explain the difference pretty well...

For example, Saudia Arabia is wealthy and extremely religious... I hope I don't have to explain the counfounding factor in this case.

Regarding the USA: even though it scores high in religious people rates, it's also one of the free-est places in the world, so people who are not believers are still relatively free to go about their daily lives without problem, unlike in most other highly religious nations. The USA also happens to have the biggest military in the world, capable of taking over and applying pressure on its enemies in order to further its economical interests... I think these two factors should be sufficient to explain quite well why the USA is an exception to the rule.

Well I've heard that in the US in certain areas (mainly rural) that if you want to get a job and are not a churchgoer you can forget about it :) If that's true it's not all that free to be unreligious in my opinion. But I've never even visited the US so I'm not sure how true this is.

In Europe this is absolutely not a thing. Well, unless you want to work at a religious organisation I guess.

Why don't you just spell it out explicitly? What you are implying certainly isn't obvious.

I was trying to "imply" that religious countries tend to be poorer than non-religious countries according to the data (why that is, I am not sure myself but probably because religions limit your society and specially creativity in ways that are damaging to economical success) and if that's not the case for a particular country, that's just because there are stronger counfounding factors that can explain the disparity... is that what you were expecting me to say???

I wasn't expecting anything, that's why I asked. What are the confounding factors for the countries you named?

I explicitly mentioned those factors for the USA already. Do you really want me to be more explicit about Saudi Arabia?? Oh well, ok: oil.

I mean, even back in your original post it would simply have been quicker to say "due to oil" than "I hope I don't have to explain the counfounding factor in this case," even without a couple more posts arguing about it.

Note taken: never assume anything to be obvious to others.

I still don't really understand how any of these are confounding factors? I'm not a statistician but as I understand it the definition of a confounder is that it has a causal relationship with both A and B, which masks the causal relationship between A and B.

A and B here are religion and wealth/technology I understand (sort of) the points you are making about wealth, but how does any of this stuff link causally with religion?

The hypothesis is: religious countries tend to be poor.

One can read this as "religion causes people to be poor", but that's not exactly the same because when a whole country is religious, its institutions and codes will tend to be based on religion rather than science or a rationalist moral system.

With this, I hope I am making it clear that saying that religion can cause a country to be poorer than it could be otherwise is not the same as saying the same about a specific person or family.

Now, we have an extremely simplistic formula based on this hypothesis: the more religious a country's population is, the poorer the country should be.

But many factors can influence a country's wealth, including its military power (which could be a result of its priorities, not necessarily of its wealth in the first place), how much its culture/religion incentivate creativity and support new ideas (things which tend to improve technological advances which usually translate into wealth), how many natural resources it was blessed with, how honest the population is in general and so on. These are all confounding factors (they may affect the results) which can cause some countries to deviate from our original hypothesis... in a canonical scientific experiment, to know if the hypothesis is correct, you would need to find ways to remove these factors from the equation so that the only important factors that matter are those you include in your hypothesis. That can't be easily done for a study of countries because there's just not enough of them.

So the best we can do is see how well most countries fit into our hypothesis and try to understand the most important (confounding) factors that can cause countries to deviate from our hypothesis to see if the outliers can be explained away (or if they show a genuine weakness in the hypothesis) so that we can have a higher level of confidence that the hypothesis seems correct.