spain: regional income per capita + pisa scores

just making a point.

and that is that there is a very low correlation between regional differences in income and pisa scores in spain (regional income per capita index taken from here [pdf], pg. 4 – i took a simple average of the provinces in each autonomous community to get the income per capita index figures for each autonomous community. sorted by reading scores, highest to lowest, because that’s what i did last time — don’t remember why):

spain - pisa scores and income per capita 02

this is opposed to, interestingly, the strong correlation between latitude and regional differences in pisa scores in spain (see previous post for more info on sources):

spain - pisa scores and latitude

that is all! (^_^)

previously: northern vs. southern spanish iq, redux

(note: comments do not require an email. ¡olé!)


  1. Is purchasing power the same for all the regions of Spain? Sometimes just looking at per capita income can be misleading, certainly in a large country at least. Different cost of living needs different income


  2. @bleach – “Is purchasing power the same for all the regions of Spain? Sometimes just looking at per capita income can be misleading, certainly in a large country at least.”

    oh, you would bring that up! always the troublemaker. (~_^)

    you’re probably correct, though. i don’t know at all, but presumably it is cheaper in somewhere like andalusia as compared to madrid or catalonia (barcelona). i probably won’t be checking into that, though.



  3. Oh, the correlation is ~0.25; didn’t see that initially.

    That’s strangely low (though still significant). I wonder why Spain would be an exception to the usual pattern of strong correlations between PISA results and income per capita.

    One possible contributing factor – the islands perhaps shouldn’t really be included in the data set, as they are tourist based economies and as such quite “artificial” (like resource based economies).


  4. It’s not just the islands, the whole country gets a lot of tourists. According to Wikipedia it has the fourth most tourists in the world, but much more in the South. It may be that the northern part of the country has a more intelligent, and also introverted population. Extraverted people have an edge in tourism no doubt as introverts and intelligent people have in reading. But in such an extreme tourist economy it pays more to be a people person.


  5. Doesn’t latitude has “something to do” with muslim conquests of the peninsula? The north was not conquered, ehm, much. And visigoths?


  6. @krakonos – “Doesn’t latitude has ‘something to do’ with muslim conquests of the peninsula? The north was not conquered, ehm, much. And visigoths?”

    yeah, that absolutely makes sense. (^_^)


  7. @anatoly – “Oh, the correlation is ~0.25; didn’t see that initially. That’s strangely low (though still significant).”

    still significant, yes, but much lower than the correlation with latitude.


  8. Krakonos: I don’t have genetic data on that, so I can only tell what history books and tradition say. The only unconquered regions were at the NW. As a matter of fact, what are now Cataluña and Aragon, were conquered lands. And then, you’ll have to take into account how the land was repopulated after the Reconquista. Surely, the former Marca Hispanica received French immigration from the Charlemagne times and on. After the Kingdom of Aragon achieved some stability, they didn’t participate in the Reconquista, and most of the burden was assumed by the Kingdom of Castilla, including the repopulation of the reconquered lands (Valencia being an exception in this last point, perhaps). As a matter of fact, tradition has it that most of the original people of Andalucia were replaced by the conquerors, so probably Moroccan gene contribution is higher at places outside Andalucia and Extremadura. A funny fact is that Castile had to be repopulated from the North, or so they say.


  9. Being the inspirator of this post with a provocative tweet, I guess I owe some explanation. A study like this always starts with a question, and a question implies a hypothesis. In this case, the hypothesis was: “there is a correlaton between latitude and intelligence”. The scope of the study was to find out if this correlation exists.
    Indeed, a correlaton was found but with what assumptions? The first is that the PISA score is an indicator of intelligence of the pupils (I am deliberately not using IQ here, since I am convinced that IQ is a very biased, inadequate, obsolete and overrated indicator of intelligence).
    But the PISA test is meant to measure the quality of a schooling system, not the intelligence of the pupils. So, either we can demonstrate that the inventors of the PISA test missed their objectives, or we must be extremely careful in using its values to measure something that it was not meant to measure.
    But more than this, I would not have engaged in such a study if not after trying to visualize all its implications. A simple test to see if a research is worth the effort or it is just wasted time, is to first see if its initial hypothesis leads to absurd results.
    The hypothesis that “there is a correlaton between latitude and intelligence” indeed does lead to some strange results: firstly, is this valid within a Country or across Countries? If it is valid only within a Country, an explanation must be found as to why intelligence “resets” across the boundary, considering that State boundaries are quite arbitrary they have been set very recently and more often than not they do not match ethnical diversity (at least in Europe: for example, are Spanish Basques any less clever than French Basques because they live at the South of the Pyrenees? Are Italian Sud Tyrolers less clever than Austrian Tyrolers? Are the Northern French any less clever than Walloons Belgians? Or is it the opposite instead, being by assumption southern Germans the less clever of all Germans, then they must be less clever than Northern Italians?).
    If intelligence does not reset across arbitrary State borders, then we should find that Danes are more clever than Germans, Swedes more clever than Danes and Norwegians more clever than Swedes. And this beggars belief.
    I often find in my work that simplified assumptions can be supported by strong statistics, but by neglecting “minor” variables and effects, they are completely worthless, if not misleading.
    My impression is that the hypothesis that “there is a correlaton between latitude and intelligence” is taken as true a priori for this study and that the statistics is mere polishing of this given truth.
    Having said this, I am sincerely surprised by the low correlation between regional gross income and PISA scores. I am tempted to say “bad data” but I’d rather say “badly spent money” and, as one of my masters used to say: “this requires further investigation”.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s