At her blog, hbdchick has a discussion of the north-south division in Spain where she asks if there are any differences in IQ that show up. I suggested looking through the 2009 PISA results, which she did. To follow up, I decided to map out the results for Spain and Italy. For each map, the pure black is 400 and pure white is 525 (a scale I came up with before I realized that Lombardy had a 526 average science score). To start, here are the maps for Reading scores:
It is worth noting that PISA did not release the scores for three autonomous communities in Spain: Extremadura, Castile-La Mancha, and Valencia, which form a band immediately to the north of Andalusia and Murcia. I’ve made those three mostly transparent.
For Mathematics:
For Science:
Finally, this pair of maps integrates the data from all three areas, with the red channel being Reading, green Mathematics, and blue Science. As above, the lighter a region, the higher it’s scores were, but in addition, its hue reflects the areas of the test where students did comparably better. For instance, many of the provinces in northern Italy have a slight bluish hue and that’s a result of having comparably good science scores and comparably poor reading scores:
Here’s the raw data for the maps above:
| Region | Reading | Mathematics | Science |
|---|---|---|---|
|
Spain |
|||
| Andalusia | 458 | 462 | 469 |
| Aragon | 492 | 506 | 505 |
| Asturias | 492 | 494 | 502 |
| Balearic Islands | 461 | 464 | 461 |
| Basque Country | 496 | 495 | 495 |
| Canary Islands | 444 | 435 | 452 |
| Cantabria | 488 | 495 | 500 |
| Castile and Leon | 507 | 514 | 516 |
| Catalonia | 499 | 496 | 497 |
| Ceuta and Melilla | 403 | 417 | 416 |
| Galicia | 483 | 489 | 506 |
| La Rioja | 488 | 504 | 509 |
| Madrid | 499 | 496 | 508 |
| Murcia | 484 | 478 | 484 |
| Navarre | 495 | 511 | 509 |
|
Italy |
|||
| Abruzzo | 481 | 476 | 480 |
| Aosta Valley | 506 | 502 | 521 |
| Apulia | 488 | 488 | 490 |
| Basilicata | 473 | 474 | 466 |
| Bolzano | 497 | 507 | 513 |
| Calabria | 449 | 442 | 443 |
| Emilia-Romagna | 496 | 503 | 508 |
| Friuli-Venezia Giulia | 507 | 510 | 524 |
| Lazio | 474 | 473 | 482 |
| Liguria | 480 | 491 | 498 |
| Lombardy | 514 | 516 | 526 |
| Marches | 492 | 499 | 504 |
| Molise | 465 | 467 | 469 |
| Piedmont | 487 | 493 | 501 |
| Sardinia | 469 | 456 | 474 |
| Sicily | 451 | 450 | 451 |
| Trento | 500 | 514 | 523 |
| Tuscany | 487 | 493 | 500 |
| Umbria | 484 | 486 | 497 |
| Veneto | 505 | 508 | 518 |
A Wandering Learner 
Great work.
Do you propose any reason for these differences besides differing ancestral stocks?
@hail – “Do you propose any reason for these differences besides differing ancestral stocks?”
i’m thinking differing ancestral stocks. i’m also thinking differential migration in the modern period.
apparently, a helluva lot of andalusians left for the new world in the 1800s, not to mention simply for madrid and other areas further north within spain for jobs in more recent years. the ones who didn’t choose to leave for a better life might have been the ones who were not so bright.
Of course, it’s hard to give a definitive answer here, but if asked to speculate, I’ll fire away. Certainly, there may be differences in ancestral stocks and there are some differences in climate between the north and the south of Spain. Aside from the New World shipping brain drain idea, there’s also the fact that the southern section of Andalusia was governed by the Emirate of Granada for two and a half centuries after the rest of Iberia had been reconquered, and while its Moorish rulers left its population with a lovely palace fortress in Alhambra, they may have also left a genetic and cultural imprint on the population.
In support of this conjecture, I’ll point to Murcia, which is actually somewhat close to the Spanish average in its performance and had noticeably higher performance than its neighbor to the southwest. It really is a pity that Extremadura’s results aren’t shown, since many of the New World colonists came from that area of Spain.
Galicians are considered dumb in Argentina. But the figures say otherwise.
Hello Reluctant Apostate, I’m very late to this post, but I’ve recently created some maps on GDP per capita by region in Europe (it’s for an academic project), as I wanted to see a finer gradation than most such maps provide. (My maps use a 22-level color gradation, the largest I’d ever seen was 6.)
GDP per capita of course has its limits as a reliable measure for the real wealth-producing capacity of a given people, nonetheless I think it can be informative, and looking at your PISA maps for Spain and Italy I saw some possible correlations. Just in case you’d like to take a look, here’s a link:
http://www.scribd.com/doc/59601902/GDP-Per-Capita-by-Region-2006-Eurozone-With-Political-Borders-Indicated
Thanks for putting these maps together in any case, very interesting stuff.
I don’t agree with hbd chick. I am myself from the first region in all three categories: Castile and Leon, and can tell you that migrations from this region have been (and still are today) enormous. Actually I was born in the Basque country, because my father emigrated to the industrial areas of Basque Country and Catalonia, although went back to Castile and Leon when I was 7. Andalusia is famous because of the absolute numbers, but in Castile and Leon still today people move away, so migration is deeper still today, and yet: the only region in probably the whole of Europe that scores first in its own in the three categories available. What is really amazing is the sheer lack of correlation between these scores and the GDP in Castile and Leon. The answer seems to be much more complicated, but I would probably suggest the scores are correlated to the GDP divided by the actual number of immigrants in the population (meaning resources per immigrant childd): the higher this is, the lower is the PISA score. In this case it makes sense the very low scores of the Mediterranean regions plus Madrid and the relatively high scores in the Basque Country, where the immigrant population is relatively high but the education resources available are much higher.