I would like to ask what method do you use in statistic, to avoid making conclusions too fast.
For instance. Country natural resources vs poverty
Observing Nigeria for example, may lead you to conclude that the richer a country is in natural resources, the more poverty it has. Is there a method you can use to prevent this conclusion, or does this fall in range of common sense.
Not an answer, but here's an observation.
In the example that you gave you want to predict "poverty" and you chose as predictor the "amount of natural resources". Now, there's infinitely many variables that you could use to predict "poverty". My observation is that instead of formulating your question as "What method should I use to avoid making conclusions too fast", you should think about the more general question: "Given a dependent variable that I want to predict, which predictors should I include in my model?"
I don't have an answer to that. In practical terms my best approach is partly to use some common sense, like you said, and partly what data I have access to.
The way you avoid reaching conclusions too fast is by making sure that the statistician/data analyst knows something about the subject matter and is working with someone who knows it well. Then you have to make sure that both people are thinking and not "fooling themselves". As Richard Feynman said:
“Science is a way of trying not to fool yourself. The principle is that you must not fool yourself, and you are the easiest person to fool.”
For your specific example, your conclusion doesn't make any sense since you have mentioned only one country. If I wanted to try to figure out the relationship between a country's natural resources and its poverty, I would look at lots of countries. I would also include control variables that I would figure out by consulting with an expert on geography. I'd certainly include one for climate, a couple for form of government, probably one for income inequality and more.
