5
$\begingroup$

Recent questions like Why are the 10 year German bunds and 10 year US treasury yield highly correlated? and Baht drops and Gas Prices too? belong to a certain class of question that asks why does thing one correlated with thing two when I would expect it to be very/slightly different level of correlation. These questions are usually low-effort to ask and are essentially unanswerable or answers would often be opinion based. I'd like to see if there is consensus against these sorts of questions.

However, somewhat somewhat similar questions are of good quality: Correlation between salary level and housing prices in a town and Can econometrics test for correlation or causality between prices and corruption?. I'm not sure how to discard the former without junking the latter.

Do you share my dislike of these questions? Can anyone suggest a limiting principle or good distinguishing feature for these correlation questions?

$\endgroup$
9
$\begingroup$

As you yourself mentioned, in general these type of questions cannot be discarded. I would say that the criterion rests again with the question's effort:

If the question is "Why X is correlated with Y (in country-A), since they shouldn't, at least according to theory-Z (which summarily argues that...) (AND/OR) empirical evidence from country-B indicates that they are not?" it shows that the OP has devoted some time and thinking on the matter, and also, it brings potentially useful information in the site.

If the question is "Why X is correlated with Y (in country-A)" /question-mark /end of post, then it is the classic 5-year-old phase, where a child asks unstoppably "why this and why that?" -and although the site's policy does accommodate undergraduate questions, I don't think it can be extended to accommodate primary-education-level questions.

$\endgroup$
  • $\begingroup$ +1 for the last sentence. Hehe $\endgroup$ – Ubiquitous Aug 22 '15 at 8:10
5
$\begingroup$

Any question that asks "Why does X correlate with Y", could be required to answer "Why shouldn't it".

$\endgroup$
  • $\begingroup$ Yeah... "this is because this isn't not" or "it is because it is" or "I am because I am, so shut up" ;) I thought that economy is meant to (or at least attempt to) provide scientific like answers. Well I think some of such questions actually might matter. A person asking such a question is probably not interested in a literal or superficial explanation, but about deeper undergoing connections between two or more factors - provided that there are such significant correlations, or at least to read an answer stating sth like "there is no significant correlation between both". $\endgroup$ – forsberg Aug 29 '15 at 11:34
  • $\begingroup$ @forsberg My point has essentially the same motivation as Alecos' explains, so read his answer (which came later) for clarification. $\endgroup$ – FooBar Aug 29 '15 at 12:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .