Richard Lynn and I have just written a paper which we submitted to a journal. You can see a draft here: https://drive.google.com/file/d/0B7hcznd4DKKQS0VwZHROYTRUN2M/edit?usp=sharing

Is there direct evidence that the problem-solving subtest is less affected by education than the crystallized subtests? As the Flynn effect has shown, it's hard to estimate cultural dependence of tests a priori.

I am not arguing that fluid intelligence is less dependent on culture, but on a subspecies of culture, that is schooling. PISA Math, Science and Reading is not even just a general "crystallized intelligence" test, it's a test that SPECIFICALLY assess skills acquired through schooling, because it was created with that purpose. PISA creative problem solving assesses reasoning and problem solving abilities that do not require knowledge acquired through schooling. It's not the Raven's where you can argue for test-taking skills, but it's a completely new test based on tasks not commonly used in more common psychometric tests.Here you can see a list with examples of items used in PISA creative problem solving: http://www.keepeek.com/Digital-Asset-Management/oecd/education/pisa-2012-results-skills-for-life-volume-v_9789264208070-en#page52

I got the reviews for this paper and publish them here, according to my policy. Reviewers 1 and 3 had no objections

Reviewer #2

2. "I am not sure that "creative problem solving" from PISA is a "pure" measure of fluid intelligence. "Have I ever stated that this is a “pure” measure of fluid intelligence? No. A Nobel prize shall be awarded to who finds a pure measure of fluid intelligence.
PISA creative problem solving is as good a measure of fluid intelligence as any, Raven’s included. All measures of fluid intelligence require some knowledge to be solved. No measure of fluid intelligence is a pure measure of g, even Raven’s (which are commonly used as THE test of fluid intelligence) have a g loading far less than 1 (about 0.7-0.8). Thus this comment is irrelevant.
3. “The author made different comparisons” because data for individual regions were not availabale. PISA creative problem solving scores are published only for the 5 macro-regions.
4.” it seems to me that the magnitude of the difference is still high but it seems to be due to the extremely high performances of the north and to a performance about the average in the south”.
Yes, and this is precisely why we have stated that Italy (including southern Italy) performs relatively better on fluid intelligence tasks. This reduces the disadvantage of the south compared to the rest of Europe and increases the advantage of the north, by about 1-2 IQ points in each direction.
5. “The assumption that the difference between creative problem solving and the other measures reflects the contribution of the school system and this is valid to the same extent for all the Italian regions does not seem sufficiently justified.” It’s just a very plausible assumption since knowledge required by PISA reading,math and science is learned in school, whereas PISA Creative Problem Solving was created with the specific purpose of testing abilities not requiring scholastic knowledge. Do not forget PISA reading, math and science is not just a test of crystallized intelligence, it’s a test of scholastic intelligence, creating with the specific purpose of assessing scholastic aptitude.
6. ""it can be inferred that not only the difference in fluid intelligence but also that in crystallized intelligence between southern and northern Italian students, cannot be explained by teaching quality and differences in resources between northern and southern schools" This is only a speculation, and in fact the author does not have sufficient evidence on this point. " To the extent that the abilities assessed by PISA Creative Problem Solving are less dependent on schooling compared to PISA reading, math,science it can be inferred that the difference between north and south is not dependent on schooling. Otherwise, if the difference were dependent on schooling, the tests more sensitive to it (PISA math, reading, science) would show a bigger north-south differential. This logic is similar to Jensen’s method of correlated vectors. As stated in point 5, it is a very plausible assumption that PISA Creative Problem Solving is less dependent on schooling, and if the assumption is valid, then this the logical conclusion is that the difference between north and south cannot be entirely explained by schooling.
7. “The final speculations on migration and genes are not convincing. Furthermore , if North-Western Italy is more interested from migrations from Southern Italy and Africa, the effects should be specific and produce a lower performance in the case of this area”.
We have discussed at length why the north-west, despite being the recipient of southern immigrants, could have scores as high as the north-east. This could be due to selective migration. Besides there was also migration from the north-east to the cities in the north-west, and this could have lowered the IQ of the North East, whilst increasing the IQ of the North-West, thus compensating for the IQ lowering effect caused by the flow of southern immigrants. No official figures are available but it’s likely that the percent of south Italian ancestry in the North West is about 20%, as can be inferred from surnames and migration numbers. Even if the difference between north and south Italy were entirely genetic (and we’re arguing that it’s at least partially but not totally genetic),a 20% of south Italian ancestry would have an IQ lowering effect of only 2 points, which could be masked by other factors such as selective migration from other parts of Italy to the North West or cultural effect.
The reviewer also says that the North-West is the recipient of more African immigrants. It’s indeed a pity that scores are not available for Italian Natives vs Immigrants. The presence of lower IQ immigrants is going to mask the true IQ difference between northern and southern natives, because (being more different from it) it weighs more on the northern average, thus taking down the average by a greater extent. Thus the actual score difference between north and south natives is likely higher than 10 points.

Concerning 5° (and 6° by extension) I can easily conceive the idea that knowledge helps to develop creativity, although i doubt it will have big impact. Where is 1° by the way ?

Concerning 5° (and 6° by extension) I can easily conceive the idea that knowledge helps to develop creativity, although i doubt it will have big impact. Where is 1° by the way ?

The name "creative problem solving" is misleading. It's novel problem solving, which is roughly equal with fluid intelligence. The concept of creativity in psychology is ill defined.

Question 1 is being dealt with by my co-author (who is familiar with the history of this debate, which he started).

Richard Lynn and I have just written a paper which we submitted to a journal. You can see a draft here: https://drive.google.com/file/d/0B7hcznd4DKKQS0VwZHROYTRUN2M/edit?usp=sharing

The Italian cognitive ability differences go pretty far back, no? See, for example: "Uniting Souls and Numeracy Skills. Age Heaping in the First Italian National Censuses, 1861-1881"
http://www.uv.es/fresh2013/Programme_files/session%201_1.pdf

Also, table 6 here:
http://tobias-lib.uni-tuebingen.de/volltexte/2009/3780/pdf/Dissertation_final.pdf

You should be able to conduct a cross lagged analysis to provide evidence for your implied cognitive ability --> school quality causal model.
http://srmo.sagepub.com/view/the-sage-encyclopedia-of-social-science-research-methods/n202.xml

Emil and I were going to try this using Age Heaping scores on the international level. I think we were waiting for him to finish up his data set and for Meng Hu to help us look into method.

Thanks for sharing the link on Italian numeracy skills. Here you can see the fluid g (Creative Problem Solving) scores for countries sampled by PISA 2012: https://docs.google.com/spreadsheets/d/16vdyElN4nwnfk0J6IKXq6Zf6RMdqbUUXKxYKgkxTW9A/edit?usp=sharing

The paper has finally been published in Intelligence. Whoever wants to download the final version can do so from here:
https://drive.google.com/file/d/0B7hcznd4DKKQVldnaGV4cVAwYW8/edit?usp=sharing

Richard Lynn's nonsense has already been refuted by a lot of evidence (including by Templer 2012, despite what you claim in your study):

http://italianthro.blogspot.com/search/label/IQ

And even though PISA Creative Problem Solving is still not a real IQ test, that data refutes Lynn too (again, despite what you claim). All Italian regions performed better on it than on the regular PISA tests (and "Italy South Islands" showed the biggest improvement), while many higher performing "Northern" countries either improved a lot less or actually did worse. It seems the closer we get to "pure g", the smaller the "IQ gap" becomes.

Also, "Eurogenes" is not peer-reviewed research with verified samples and proper interpretations of data. It's just some amateur playing on his computer. According to a very large recent study, the Mediterranean Sea has acted as a barrier to gene flow through isolation, and there's been no significant Middle Eastern or North African admixture in any part of Italy since Neolithic farmers' original settlement of Europe:

http://italianthro.blogspot.com/2014/06/mediterranean-sea-as-genetic-barrier.html

I guess you and your buddy Lynn didn't notice that based on the new better test, Southern Italy's "IQ" is only 2 points lower than Sweden and Denmark's. Those Nordic countries must have almost as much of that filthy brain-damaging MENA admixture, while Northern Italy must be populated entirely by pure Aryan Übermenschen.

And even though PISA Creative Problem Solving is still not a real IQ test, that data refutes Lynn too (again, despite what you claim). All Italian regions performed better on it than on the regular PISA tests (and "Italy South Islands" showed the biggest improvement), while many higher performing "Northern" countries either improved a lot less or actually did worse. It seems the closer we get to "pure g", the smaller the "IQ gap" becomes.

As we stated in the paper, the PISA CPS vs PISA M,S;R improvement amounts to only 1 IQ, a very small improvement.
Northern Italy has received a lot more low achieving immigrants than southern Italy from 2000 to 2012, yet its IQ advantage is still there.
The IQ gap between north and south Italy is essentially the same in PISA CPS and the other PISA tests, so your claim that the closer we get to pure g, the smaller the IQ gap becomes is groundless. Given that the proportion of immigrants is much higher in the north and immigrants have lower PISA scores (check PISA documents for immigrants), the difference between northern and southern natives has to be bigger.
INVALSI is not a measure of g but simply of mastering school curriculum and the difference between north and south is much smaller, so this goes against your contention.

Also, "Eurogenes" is not peer-reviewed research with verified samples and proper interpretations of data. It's just some amateur playing on his computer. According to a very large recent study, the Mediterranean Sea has acted as a barrier to gene flow through isolation, and there's been no significant Middle Eastern or North African admixture in any part of Italy since Neolithic farmers' original settlement of Europe:

http://italianthro.blogspot.com/2014/06/mediterranean-sea-as-genetic-barrier.html

Eurogenes K13 East Med component reflects immigration from the Middle East dating to Neolithic farmers. Nobody claimed that it reflects historical migration. This is simply your assumption.

I guess you and your buddy Lynn didn't notice that based on the new better test, Southern Italy's "IQ" is only 2 points lower than Sweden and Denmark's. Those Nordic countries must have almost as much of that filthy brain-damaging MENA admixture, while Northern Italy must be populated entirely by pure Aryan Übermenschen.

South/Islands IQ is for some odd reason higher than Southern Italy (95.15 vs 93.28) probably because Sardinia has a higher IQ than Sicily.The difference between South Italy and Denmark is around 4 IQ points, but it's a bit higher if we include only natives.

The IQ gap between north and south Italy is essentially the same in PISA CPS and the other PISA tests, so your claim that the closer we get to pure g, the smaller the IQ gap becomes is groundless.

INVALSI is not a measure of g but simply of mastering school curriculum and the difference between north and south is much smaller, so this goes against your contention.

I was talking only about the different PISA tests in the whole European context, not just Italy. But yes, the INVALSI data proves how inaccurate "IQs" calculated from PISA scores are. None of those tests are legitimate intelligence tests, not even the CPS.

Based on Ravens and CAS, D'Amico et al. found no significant North-South differences in Italy:

http://italianthro.blogspot.com/2012/03/rebuttal-to-richard-lynns-reply.html

Eurogenes K13 East Med component reflects immigration from the Middle East dating to Neolithic farmers. Nobody claimed that it reflects historical migration. This is simply your assumption.

Your new study doesn't mention Neolithic farmers anywhere, but it does say "historical times" on page 2. And the old Lynn study you're trying to confirm claims the admixture comes from the Phoenicians, Carthaginians and Arabs. So no, it's not "simply my assumption".

And either way, it's an idiotic argument because Phoenicians, Carthaginians and Medieval Arabs were highly civilized people, and Neolithic farmers were much more advanced than native European hunter-gatherers.

The difference between South Italy and Denmark is around 4 IQ points, but it's a bit higher if we include only natives.

It's more like 3 points between all Southern Italians and non-immigrant Danes, and half that if you use Swedes and only "Italy South Islands". But even 4 points wouldn't come close to proving your case. According to the Lynnian Nordic Supremacist doctrine that you support, Scandinavian IQs should not be intermediate between Northern and Southern Italian IQs. They should be much higher.

And do you seriously believe that Bulgarians, Hungarians, Slovaks, Poles and Former Yugoslavs have IQs of 82-94? Of course they don't, and neither do Spaniards, Portuguese and Greeks. Based on real intelligence tests, all Europeans have IQs of ~98 or higher:

http://dienekes.awardspace.com/articles/greekiq/

I was talking only about the different PISA tests in the whole European context, not just Italy. But yes, the INVALSI data proves how inaccurate "IQs" calculated from PISA scores are. None of those tests are legitimate intelligence tests, not even the CPS.

PISA CPS and PISA correlate very highly with IQ at the national level (around 0.9). The PISA CPS items are very similar to those found in some IQ tests.

Based on Ravens and CAS, D'Amico et al. found no significant North-South differences in Italy:

http://italianthro.blogspot.com/2012/03/rebuttal-to-richard-lynns-reply.html

I think Lynn has already showed that the Colored progressive Matrices Italian sample is not reliable.
Templer reported results from an Internet IQ test (which I had personally found and downloaded), which showed about 10 points difference between N and S Italy. The items of the IQ tests included logic and spatial problems.

Your new study doesn't mention Neolithic farmers anywhere, but it does say "historical times" on page 2. And the old Lynn study you're trying to confirm claims the admixture comes from the Phoenicians, Carthaginians and Arabs. So no, it's not "simply my assumption".

No I didn't write that. It's simply a quotation from a previous paper by Lynn. "Lynn (2010a) argued further that the north–south IQ
differences in Italy are attributable in part to immigration from North Africa and the Middle East in the more southerly regions in historical times,"
The East Med component probably represents Middle Eastern genes mostly from Neolithic farmers with some later input in historic times.
In your blog you wrote

Naglieri et al. (submitted for publication) studied the differences between the psychometric qualities of the CAS [Cognitive Assessment System] for the Italian and US standardization samples.

That post is 2 years old. Where are the datasets and the paper from Naglieri et al? Can you attach it?

I managed to download the Naglieri et al. (2013, Psychological Assessment, 25: 157-166) paper which ItalianAnthro purports to show that there is no difference between North and South Italy.
However, it can easily be seen that the comparison has little value because it is based on a very small sample size (only 55 and 121 for North and South, respectively). Such a small sample would make it difficult to obtain a representative sample in terms of socioeconomic status. Indeed, when a school director gave approval, all students in a class had to take the test. So a sample of 55 students would include only about 3 schools.
A total of 984 cases were obtained: 809 cases for the normative sample and the remaining subjects for research studies.
The sample was mainly selected from the area of Tuscany, with some portion from other sections of Italy (north, central, and
southern). The sample (see Tables 1 and 3) included participants attending public schools from central
(78.2%) and, in smaller numbers, from northern (6.8%) and southern (15.0%) Italy.
As shown in table 3, N was 55 for the North, 633 for Central and 121 for Southern. Scores: Central= 100.5 (SD 13.2); Northern= 101.2 (SD 11.9): Southern= 103.1 (SD 11.6).
With such a disparity in sample size between regions, it's possible that the authors were less careful in getting a representative sample from the regions with smaller samples.
Nonetheless the results are interesting and await replication, with bigger and more representative samples.

The scores pattern is also messed up. Central Italy always scores between North and South, in all IQ or achievement tests (PISA, Internet IQ tests, etc.) or other measures of economic performance (GDP, life expectancy, etc.) and genetically too. So it is fishy that its average score is lower than Northern and Southern Italy. Most likely more care was taken to get a representative sample for Central Italy, as sample size is much bigger (78% of entire sample) and the study had a focus on Tuscany (a region of central Italy). It's likely the authors neglected good normative sampling standards when dealing with North and Southern Italy, besides sample size being too small to have a decent power.

PISA CPS and PISA correlate very highly with IQ at the national level (around 0.9).

Not really. I provided real IQ data showing that all Europeans have IQs of ~98 or higher. Your PISA data shows some of those same Europeans to have "IQs" of 82-94. Do you believe that's accurate?

I think Lynn has already showed that the Colored progressive Matrices Italian sample is not reliable.

No he hasn't. He tried to discredit the CPM data from Cornoldi et al. (2010) by providing new data of his own, but D'Amico et al. (2012) refuted his methods and showed with more new data that Cornoldi was right:

"Lynn (2010b) uses data from several studies on Raven's test (Pruneti, 1985; Pruneti, Fenu, Freschi, & Rota, 1996; Tesi & Young, 1962) and Cattell Culture Fair test (Buj, 1981; Pace & Sprini, 1998). None of the studies used the same age groups and none were aimed at comparing IQs across regions of Italy.

Moreover, Lynn (2010b) did not consider the calculation of IQs made by the authors, but rather he recalculated the IQ scores in light of the well known and controversial (Colom, Lluis-Font & Andrés-Pueyo, 2005) Flynn effect (2007), described as a general increase of intelligence scores over the world in the last 50 years. So, for instance, an IQ of 99 collected in 1960, was increased by 4 points considering the Flynn effect = 4 of the Italian IQ in the years 1960-79.

Such procedure is questionable, as also Hagan, Drogin, and Guilmette (2008) pointed out. Indeed, different studies demonstrated that the Flynn effect is concentrated in the lower half of the normal distribution or in undeveloped countries (Colom et al., 2005), whereas a possible stagnation of IQ scores in developed ones is currently under debate (Teasdale & Owen, 2005; 2008).

[...]

Despite the minor differences between the studies, our results demonstrate quite clearly that raw scores [on Raven's Coloured Progressive Matrices] of children from Sicily are not lower than those [of children from the North and Central-South] reported by Cornoldi et al. (2010). On the contrary, they are sometimes higher."

Templer reported results from an Internet IQ test (which I had personally found and downloaded), which showed about 10 points difference between N and S Italy.

That was Lynn (again), and D'Amico et al. say what anyone can already figure out about data collected from a self-administered online test:

"In his more recent paper, Lynn (2010b) reports further evidence of the lower IQs of southern Italians. The first is the report of an intelligence test given to a sample of 50,000 individuals who self-administered the test over the internet on www.sitozero.it. This is a commercial site with an inadequate description of the psychological tests used, with a considerable amount of advertisements and without any control of scientific and methodological issues. We do not consider these non-scientific data to be suitable for making assumptions about IQs."

No I didn't write that. It's simply a quotation from a previous paper by Lynn. "Lynn (2010a) argued further that the north–south IQ
differences in Italy are attributable in part to immigration from North Africa and the Middle East in the more southerly regions in historical times,"
The East Med component probably represents Middle Eastern genes mostly from Neolithic farmers with some later input in historic times.

It's a quotation from the previous Lynn paper that you're claiming to support with your new paper. And Lynn is the co-author of that new paper. Stop telling ridiculous lies.

And what evidence do you have that those groups (Neolithic farmers, Phoenicians etc.) were unintelligent and would have lowered people's IQs? None.

The scores pattern is also messed up. Central Italy always scores between North and South, in all IQ or achievement tests (PISA, Internet IQ tests, etc.) or other measures of economic performance (GDP, life expectancy, etc.) and genetically too. So it is fishy that its average score is lower than Northern and Southern Italy.

The score pattern is even more messed up with your PISA data. If "Northern European" Ancestry = Higher IQ and "Middle Eastern" Ancestry = Lower IQ (as you and Lynn believe) then how can Danish and Swedish IQs be in-between Northern and Southern Italian IQs? Isn't that "fishy"? Or are Padanians magically more Nordic than real Nordics? How amazing!

The score pattern is even more messed up with your PISA data. If "Northern European" Ancestry = Higher IQ and "Middle Eastern" Ancestry = Lower IQ (as you and Lynn believe) then how can Danish and Swedish IQs be in-between Northern and Southern Italian IQs? Isn't that "fishy"? Or are Padanians magically more Nordic than real Nordics? How amazing!

The IQ of Northern Italians is really around 101 if we average scores on PISA tests. This is only 1 point higher than UK, Germany, France Native populations. A population's intelligence is not only the result of admixtures due to migrations, but also INTERNAL dynamics, such as dysgenic trends or higher darwininan fitness of more intelligent individuals. As the genetic components reflect mostly migrations dating to before Classical Antiquity, in the last 2000 years there has been room for genetic improvement of some countries or regions over others. Given the relative isolation experienced by North and South Italy up until the unification of Italy, this is not impossible. In fact, Piedmont (Piemonte) is the region with the lowest IQ in "Padania" (the north proper), scoring lower than Lombardy, Trentino, Friuli, Veneto. This is likely due to its massive admixture with southern populations, due to the massive migrations from Sicily, Puglia, Campania to Piemonte after the Great War. The southern admixture in Piemonte is probably around 1/3.Accordingly, its IQ is about 2 points lower than the rest of the north. This would translate into a real genotypic difference of about 2*3= 6 IQ points, which after accounting for environment is probably the real difference between North and South.
I have administered the CRT, an intelligence-critical thinking test based on 3 items, and found differences between North and South, not big ones (due to restriction of range and ceiling effects). d is only about 0.16, equivalent to an IQ advantage of 2.5 points for the North. However, after correcting for unreliability, restriction of range, ceiling effects, etc. the difference is going to be greater.
Results of t-test can be found here:

https://docs.google.com/document/d/1g1B24K7RPGngYl_KgheAUVD8YWTLvQrddeRlGR7XsiI/edit?usp=sharing