I have changed the title of my paper to "Sexual selection as an evolutionary mechanism behind sex and population differences in fluid intelligence". I will delete the word IQ from my paper as it has created confusion and the wrong impression that crystallized tests could be used to test my hypothesis. What I hate the most about science is misunderstanding one's ideas, and I'd rather not publish my work if it means going through harsh rounds of rudeness and incomprehension. I will not respond to critiques leveled at my thesis on the grounds of not having adopted measures such as PISA reading (and perhaps math), or other bad proxies for fluid g (such as age heaping) because my hypothesis has nothing to do with these abilities.

"Here is what Wikipedia says..."

Notice that the source of this passage is just another Wikipedia page in another language. No matter. It looks like we agree to change "intersexual competition" to "intersexual selection." That is a much less confusing term.

Ok we agree on this. I've made the corrections you suggested and added a disclaimer at the end. New paper is attached.

P.S.: As I said in a previous post, my hypothesis regards fluid intelligence. This is why I changed the title and I replaced IQ with g or gf where appropriate, to avoid generating confusion in the reader.

You shouldn't threaten to not approve of my paper. Your approval or not does not make my thesis right or wrong.

As you are so hostile to my commentary -- despite having personally requested my input in the first place -- I will accept your invitation to cease reviewing.

Before moving on, I will note, though, that Lynn and Irwing's 2004 and 2005 Raven's matrices data also seemed not to show a (positive) r (national IQ x M/F difference). Moreover, while it
has been said that PIAAC's Problem Solving test is "more related to" fluid intelligence, this test also failed to show the predicted national IQ x M/F diff association. Granted these indexes are imperfect: Lynn's data is unrepresentative; the PIAAC PS N is only 18 to 19, depending on how cases are counted. But it's not as if you found strong support from PISA CPS in the first place.

In general, you haven't established your explanandum. I would start there.

You shouldn't threaten to not approve of my paper. Your approval or not does not make my thesis right or wrong.

As you are so hostile to my commentary -- despite having personally requested my input in the first place -- I will accept your invitation to cease reviewing.

Before moving on, I will note, though, that Lynn and Irwing's 2004 and 2005 Raven's matrices data also seemed not to show a (positive) r (national IQ x M/F difference). Moreover, while it
has been said that PIAAC's Problem Solving test is "more related to" fluid intelligence, this test also failed to show the predicted national IQ x M/F diff association. Granted these indexes are imperfect: Lynn's data is unrepresentative; the PIAAC PS N is only 18 to 19, depending on how cases are counted. But it's not as if you found strong support from PISA CPS in the first place.

In general, you haven't established your explanandum. I would start there.

You totally misunderstood my thesis. It's also partly my fault because I used the word IQ when I should have used the word fluid intelligence or gf throughout the paper, but I made it clear in the introduction and methods section that CPS measures fluid intelligence, and I had already replied to previous comments from another user asking me why I had not used PISA math,science,reading explaining why I was not gonna use these and you ignored my explanation, either because you didn't read it or because you didn't want to consider them, although a reviewer should always check previous comments in the thread and it's rude to dismiss an author's explanation. My hypothesis is not about scholastic apititude, because wide access to schooling is a recent phenomenon and there has not been enough time for selection to act on such skills; also, these are affected by effort and motivation; they are not a good measure of fluid g,particularly when looking at sex differences because females outperform males on reading, whereas males outperform females on math; I could continue,etc.. The samples you cited in the last post are not representative. An analysis of sex differences requires nationally representative samples and not estimates based on a bunch of different studies.
I admit my analysis needs replication but this will only be possible in 2015 when new CPS data will be released. I want to provide a theoretical framework and predictions that will have to be tested again with new representative data. It would probably add useful data to my analysis if someone could point out to me studies on sex differences in fluid g among Blacks, Whites and East Asians, to check whether the racial pattern confirms the country-level pattern.

It would probably add useful data to my analysis if someone could point out to me studies on sex differences in fluid g among Blacks, Whites and East Asians, to check whether the racial pattern confirms the country-level pattern.

Also you said:

Gross Domestic Product (World Bank, 2014) and purchasing power per capita (GDP (PPP)) was used as an independent variable due to its potential relationship with g and sex differences in country scores. That is, GDP is known to be positively related to country IQ (Lynn and Vanhanen, 2006, 2012; Rindermann, 2012; Sailer, 2012) and could predict sex difference in g, possibly with more economically developed countries showing lower sex difference (lower male advantage) due to females enjoying more enriched environments, since rich countries probably undertake efforts to raise female schooling and intelligence in an attempt to make men and women more equal.

Controlling for GDP to control for sex equality is also questionable. The correlation between your GDP variable and the UN's gender equality index was only ~ 0.3 to 0.5 depending on how missing data was handled -- granted that when gender equality is partialed out, the national g x sex difference correlation is somewhat more robust.

Also you said:

Gross Domestic Product (World Bank, 2014) and purchasing power per capita (GDP (PPP)) was used as an independent variable due to its potential relationship with g and sex differences in country scores. That is, GDP is known to be positively related to country IQ (Lynn and Vanhanen, 2006, 2012; Rindermann, 2012; Sailer, 2012) and could predict sex difference in g, possibly with more economically developed countries showing lower sex difference (lower male advantage) due to females enjoying more enriched environments, since rich countries probably undertake efforts to raise female schooling and intelligence in an attempt to make men and women more equal.

Controlling for GDP to control for sex equality is also questionable. The correlation between your GDP variable and the UN's gender equality index was only ~ 0.3 to 0.5 depending on how missing data was handled -- granted that when gender equality is partialed out, the national g x sex difference correlation is somewhat more robust.

In general I prefer "hard" indexes based on real data (such as GDP), gender equality index is based on soft data and that is why I didn't use it. Nonetheless, I may include it in my analysis, although I doubt it would dramatically alter the results. Chuck, you seem pretty knowledgeable about racial data so you may have data on sex differences among different races? It'd be a good test of my hypothesis.

In the abstract, you type "dimorpshism" and in your table 1, you type "Shangai-China". These should be corrected.

A note of caution regarding this finding is necessary, as phenotypic variation is correlated to genetic variation but environmental noise due to population stratification (SES, ethnicity, etc.) can easily attenuate the genetic signal.

Concerning that problem, in general, behavior geneticists partial out the "race" variable. To this purpose they can use some sort of fixed effect approach. See here for how to do that in Stata. Very straightforward.
http://www.stata.com/support/faqs/statistics/xtreg-and-effects/

As mentioned in the above link, you can use dummy variables. In SPSS, here's the procedure:
https://www.youtube.com/watch?v=R0qc4rzr9ik

In your case, for instance, you can create dummy for european, arab, asian, african countries. You can end up with perhaps 4 or more dummies, (e.g., race1, race2, race3, race4, etc., all coded 1 if the country corresponds to the race category label and 0 for otherwise), and you should include all of them in the regression equation. (If you do that, your program will probably drop one variable due to collinearity or redundancy, and the variable dropped is treated as the reference category.)

See below (I assume here that "race" variable has been created already and is a numeric variable for which the countries have been categorized into the race category to which they should belong, e.g. white countries=1, arab countries=2, asian countries=3 etc...; I assume here you have 6 race categories, but there may be more, or less, probably... so that if you have 4 races, you should remove all the numbers '5' and '6' in the series shown below) :

RECODE race (1=1) INTO race_cat0.
RECODE race (2=1) (3,4,5,6,1=0) INTO race_cat1.
RECODE race (3=1) (2,4,5,6,1=0) INTO race_cat2.
RECODE race (4=1) (2,3,5,6,1=0) INTO race_cat3.
RECODE race (5=1) (2,3,4,6,1=0) INTO race_cat4.
RECODE race (6=1) (2,3,4,5,1=0) INTO race_cat5.

REGRESSION
/DESCRIPTIVES MEAN STDDEV CORR SIG N
/MISSING LISTWISE
/STATISTICS COEFF OUTS R ANOVA CHANGE ZPP
/CRITERIA=PIN(.05) POUT(.10)
/NOORIGIN
/DEPENDENT PISA_Score_Total
/METHOD=ENTER Difference GDP
/METHOD=ENTER race_cat1 to race_cat5
/PARTIALPLOT ALL
/SCATTERPLOT=(*ZRESID ,*ZPRED)
/RESIDUALS DURBIN HISTOGRAM(ZRESID) NORMPROB(ZRESID)
/CASEWISE PLOT(ZRESID) OUTLIERS(3)
/SAVE MAHAL COOK LEVER PRED RESID.

REGRESSION
/DESCRIPTIVES MEAN STDDEV CORR SIG N
/MISSING LISTWISE
/STATISTICS COEFF OUTS R ANOVA CHANGE ZPP
/CRITERIA=PIN(.05) POUT(.10)
/NOORIGIN
/DEPENDENT PISA_Score_Total
/METHOD=ENTER Difference GDP
/METHOD=ENTER race2 race3 race4 race5 race6 race1
/PARTIALPLOT ALL
/SCATTERPLOT=(*ZRESID ,*ZPRED)
/RESIDUALS DURBIN HISTOGRAM(ZRESID) NORMPROB(ZRESID)
/CASEWISE PLOT(ZRESID) OUTLIERS(3)
/SAVE MAHAL COOK LEVER PRED RESID.

(edit: I forget to precise, those are two different ways to perform this kind of fixed-effect regressions but they produce identical results, at least for the coefficients of variables other than the dummies which have coefficients differing only in function to which category is the reference variable.) Normally in logistic regression, what happens is that race2 to race6 have coefficients expressed in relation to race1 (reference category because entered last) but the point of interest here should be how they affect the other indep variables. It seems as far as I see that it works differently in linear regression, I don't know why but if the other independent var are unchanged, it's not necessarily a problem.

Can you do this analysis ? If you think it's useful of course (personally I think it's useful otherwise i woudn't bother).

Besides, do you have more information about PISA (CPS) ? You say it's like fluid intelligence, but it still is an achievement test. Does it really looks like the fluid subtests in most IQ tests ? (specifically, I'm talking about the items/questionnaires.)

In the abstract, you type "dimorpshism" and in your table 1, you type "Shangai-China". These should be corrected.

Can you do this analysis ? If you think it's useful of course (personally I think it's useful otherwise i woudn't bother).

Besides, do you have more information about PISA ? You say it's like fluid intelligence, but it still is an achievement test. Does it really looks like the fluid subtests in most IQ tests ? (specifically, I'm talking about the items/questionnaires.)

I'll fix the errors.Regarding race, I think doing that analysis would create more problems than it solves. I've explained in the paper that PISA is not an achievement test, instead it taps into fluid intelligence...the explanation is in the Methods section. If that is not enough, you can find example items from the CPS here: http://www.keepeek.com/Digital-Asset-Management/oecd/education/pisa-2012-results-skills-for-life-volume-v_9789264208070-en#page37
If you are satisfied with the paper, you can give your opinion concerning approving or not publication.
It's not really an achievement test (and it was built with the specific aim of not testing achievement, in contrast to PISA reading,math and science).

It's difficult for me. For the moment, I'm in full reflexion with comments made by Peter Frost and more importantly, the one made by John Fuerst. For the latter, it seems to me that whether Gf or Gc is better measure of intelligence is not clear at all. See here: "Minimally biased g-loadings of crystallized and non-crystallized abilities" (the authors say that Gc and Gf are equally important but that Gf seems to have a slightly higher loading than Gc). Of course this can still be debated but that's not the problem here. Normally, Gc and Gf should be highly correlated. If you find a correlation using Gf, there should be one with Gc, although in (much?) smaller magnitude. The fact that John has reported a correlation of just zero... I don't know how to interpret this in light of your hypothesis. Maybe you should add some comments as to why your hypothesis does not expect any relationship with crystallized (or non-fluid) measures of intelligence. Alternatively, you may want to look at the data on gender difference (across races) on fluid g. Of course if I had those data, I will give you them, but...

Perhaps an idea is to wait for the next batch of CPS data and see if it replicates? If it 1) replicates in the next CPS data, and 2) does not in the other PISA data, then I will agree that it is unlikely to be a coincidence. The current results look more like a coincidence.

Note also that the underlying theory assumes that there is sex-linked heritability of g. The meta-analysis listed by Jensen 1973 showed exactly the same r for same-sex and opposite-sex dizygotic twins. This I take it shows that there isn't any noticable sex-linked heritability.

In general, cool idea, but unconvincing data.

Coincidence? Very funny. I found a bunch of correlations that match my predictions, you guys here seem to focus only on a very simplistic interpretation of my paper (sex differences>higher CPS), but I also analyzed SD, male height data. Very hard to explain away with a coincidence. And Emil you're not even a reviewer for OBG. We've already discussed about Jensen's meta-analysis and agreed that it has limitation because ceilings restrict range which would obscure correlations, as mutations located on the X chromosome and linked to mental retardation or very high IQ have effects that will not be detected by standard, very old IQ tests from an outdated meta-analysis (1976). Furthermore, I cited evidence in my paper that intelligence genes are disproportionately located on the X chromosome, and this is bound to produce sex-linked heritabilty and this is the most likely explanation for higher male than female SD in CPS (and in general IQ) scores.

. The fact that John has reported a correlation of just zero... I don't know how to interpret this in light of your hypothesis. Maybe you should add some comments as to why your hypothesis does not expect any relationship with crystallized (or non-fluid) measures of intelligence. Alternatively, you may want to look at the data on gender difference (across races) on fluid g. Of course if I had those data, I will give you them, but...

You guys are forgetting that my analysis is not focused exclusively on the relationship between CPS score and sex difference, but also on others: correlation between variance and sex differences, which is negative matching the prediction that sexual selection reduces phenotypic variation.Also the difference between male and female variance (Male SD-Female SD) is inversely correlated to sex difference, matching prediction that selection reduces variance more in the selected sex (male).
I also found a negative correlation between male height and CPS, matching results from my analysis of genetic polymorphisms for height and IQ.
I am open to the challenge of testing my hypothesis on racial data, and see if the pattern holds for East Asians, Blacks and Whites (East Asians should have higher sex dimorphism, lower SD; Blacks should have lower sex dimorphism and higher SD). I was not able to find data on races broken down by sex but probably some of you guys can get access to it.It's not an impossible feat. Chuck in particular seems to know a lot of US datasets which could be amenable to this treatment.

It's not even true that the correlation reported by John is "just zero". The correlation between 2000-2009 PISA math and sex differences was r (df=71) = 0.142 ns; for reading r (df=73) = -0.075 ns!
Of course you're not gonna find a correlation between sex difference in reading and total reading score, because reading does not show a male advantage (to the contrary, it shows a female advantage and thus is not sexually selected!). The correlation between PISA 2000-2009 math and sex differences was weak but it could get stronger after controlling for GDP.Math is more appropriate candidate for sexually selected abilities, as it shows a male advantage. I will run this analysis controlling for GDP and perhaps gender inequality and report back the results.

I haven't been feeling well, so I haven't been paying attention to the discussion.

Regarding your request for race x sex data, you will only find this for a couple of countries (e.g., the U.S., South Africa, etc.) This data shows no consistent support for your hypothesis. While you are dismissive of them, I think inter-national convenience samples provide some contrary evidence. These do not show larger M/F differences in higher IQ nations. Just look at the Raven's scores in some of the sex difference papers. See also Lynn's South Africa race x sex data.

Here are some issues:

(1) There is as much, if not more, evidence against the hypothesis of sex differences in gf in general as there is for it. So your sex-selection for gf hypothesis is on shaky grounds to start. You should probably start by presenting the PISA data as new evidence for sex differences, not by assuming that there are such ones or that the PISA CPS is necessarily a good measure of gf.
(2) There seems to be no corroborating evidence for the hypothesis. You don't seem to see a (positive) national IQ x gf sex difference interaction in the reviews mentioned. You don't seem to see this on the PIACC problem solving in technology subtest.
(3) The equation of gf with g (intelligence), while popular, is dubious. Why if gf is g -- thus rendering gc a product not a source a la Cattell's theory -- does gc have a genetic architecture distinguishable from gf?
(4) The absence of national IQ x math/reading sex differences would be interesting given (a) your model and (b) proposals that math/reading sex differences resulted from sexual selection, e.g.,Geary, D. C. (2014). Evolved Sex Differences in Modern Context.

These noted:

(1) My original comment was indeed based on a misunderstanding of the issue (-- it would have helped if you had summarized your argument as requested). Nonetheless, it would be interesting to explore math/reading differences (net of g) from a (differential) sexual selection framework, but that's a separate issue.
(2) My gf counter evidence, while suggestive, is not robust.
(3) Whether gf is g is irrelevant to the issue. Math. reading, and science differences are confounded with broad factor (verbal/quantitative) sex differences. You could extract g-scores using math, reading, science, and problem solving, but you are right that this would be difficult, time consuming and generally not worth the effort.

Overall, I find you evidence -- for your explanandum, let alone your model, unpersuasive; of course, that's not good reason to not approve a paper.

[quote][/quote]

re: MH

I used:

1=West
2=Latin America
3=Middle East
4=Asia
For Gender Equality, I used China's scores for HK, Shanghai, Taiwan, and Macao and Macedonia's scores of Serbia and Montenegro.
Sex diff2 are the sex difference scores with the 4 Chinese scores averaged and condensed into "greater China".


RECODE Region (1=1) INTO Region_cat0.
RECODE Region (2=1) (3,4,1=0) INTO Region_cat1.
RECODE Region (3=1) (2,4,1=0) INTO Region_cat2.
RECODE Region (4=1) (2,3,1=0) INTO Region_cat3.
REGRESSION
/DESCRIPTIVES MEAN STDDEV CORR SIG N
/MISSING LISTWISE
/STATISTICS COEFF OUTS R ANOVA CHANGE ZPP
/CRITERIA=PIN(.05) POUT(.10)
/NOORIGIN
/DEPENDENT Scores
/METHOD=ENTER Sexdiff GenderEquality2
/METHOD=ENTER Region_cat0 to Region_cat3
/PARTIALPLOT ALL
/SCATTERPLOT=(*ZRESID ,*ZPRED)
/RESIDUALS DURBIN HISTOGRAM(ZRESID) NORMPROB(ZRESID)
/CASEWISE PLOT(ZRESID) OUTLIERS(3)
/SAVE MAHAL COOK LEVER PRED RESID.

I haven't been feeling well, so I haven't been paying attention to the discussion.

Regarding your request for race x sex data, you will only find this for a couple of countries (e.g., the U.S., South Africa, etc.) This data shows no consistent support for your hypothesis. While you are dismissive of them, I think inter-national convenience samples provide some contrary evidence. These do not show larger M/F differences in higher IQ nations. Just look at the Raven's scores in some of the sex difference papers. See also Lynn's South Africa race x sex data.

Here are some issues:

(1) There is as much, if not more, evidence against the hypothesis of sex differences in gf in general as there is for it. So your sex-selection for gf hypothesis is on shaky grounds to start. You should probably start by presenting the PISA data as new evidence for sex differences, not by assuming that there are such ones or that the PISA CPS is necessarily a good measure of gf.
(2) There seems to be no corroborating evidence for the hypothesis. You don't seem to see a (positive) national IQ x gf sex difference interaction in the reviews mentioned. You don't seem to see this on the PIACC problem solving in technology subtest.
(3) The equation of gf with g (intelligence), while popular, is dubious. Why if gf is g -- thus rendering gc a product not a source a la Cattell's theory -- does gc have a genetic architecture distinguishable from gf?
(4) The absence of national IQ x math/reading sex differences would be interesting given (a) your model and (b) proposals that math/reading sex differences resulted from sexual selection, e.g.,Geary, D. C. (2014). Evolved Sex Differences in Modern Context.

These noted:

(1) My original comment was indeed based on a misunderstanding of the issue (-- it would have helped if you had summarized your argument as requested). Nonetheless, it would be interesting to explore math/reading differences (net of g) from a (differential) sexual selection framework, but that's a separate issue.
(2) My gf counter evidence, while suggestive, is not robust.
(3) Whether gf is g is irrelevant to the issue. Math. reading, and science differences are confounded with broad factor (verbal/quantitative) sex differences. You could extract g-scores using math, reading, science, and problem solving, but you are right that this would be difficult, time consuming and generally not worth the effort.

Overall, I find you evidence -- for your explanandum, let alone your model, unpersuasive; of course, that's not good reason to not approve a paper.

Replies to issues. Issue 1) I found this paper by Nyborg (see attachment) concluding "The only two existing high quality studies verify independently, that there is a significant average g difference in male favour, thus confirming hypothesis 2."
PISA CPS corroborates Nyborg's claim.
Issue 2)"You don't seem to see a (positive) national IQ x gf sex difference interaction in the reviews mentioned. "
Is This claim based on visual inspection of the data or proper statistical analysis of the mentioned reviews, partialling out gender inequality or GDP, and also considering only samples from people older than 15? Regarding the racial data for South Africa, SES is a very strong confounding factor, which affects gender equality and differential cognitive stimulation of women and men. It cannot be claimed that the social and cultural view of gender is the same among Whites and Blacks living in the same country.
Issue 3) You've answered this "Whether gf is g is irrelevant to the issue."
Issue 4) I doubt there has been sufficient time for sexual selection (which is probably weak to start with) to act on math/reading differences given the incredibly narrow time scale (widespread access to schooling is a recent, post Industrial revolution phenomenon). But this point is not relevant to the current discussion.
Finally, I suspect that the PISA CPS data underestimates sex differences because the mean age is around 15.5 but we know that g continues increasing till at least an age of 18. See for example (Dapo and Dapo, 2012) which conclude "Results obtained in our study indicate that at ages of 12.6 and 16 effect sizes of sex difference in performance on tests of fluid intelligence were small. At age of 17.2 boys scored almost one standard deviation higher than girls." (PAID, vol.53; pp.811-815).

(1)

I found this paper by Nyborg (see attachment) concluding "The only two existing high quality studies verify independently, that there is a significant average g difference in male favour, thus confirming hypothesis 2."PISA CPS corroborates Nyborg's claim.

Nyborg wrote that 10 years ago! There has been a flood of papers on this since. See some of the results in table 1 of the attached. But yes you could present the CPS results as support for such differences. My point was that you shouldn't assume that Lynn and Nyborg are correct. (I imagine that a full fledged meta-analysis would turn up something since findings are either for no differences or for a male advantage, but rarely for a female one.)

(2)

Issue 2)"You don't seem to see a (positive) national IQ x gf sex difference interaction in the reviews mentioned. " Is This claim based on visual inspection of the data or proper statistical analysis of the mentioned reviews, partialling out gender inequality or GDP, and also considering only samples from people older than 15?

What do you think? Thus, I qualified my statement with "seems". I imagine that if I conducted a "proper statistical analysis" you would end up dismissing it, if the results contradicted your position.

(3)

Regarding the racial data for South Africa, SES is a very strong confounding factor, which affects gender equality and differential cognitive stimulation of women and men. It cannot be claimed that the social and cultural view of gender is the same among Whites and Blacks living in the same country.

You originally said: "you may have data on sex differences among different races? It'd be a good test of my hypothesis."

So now, on second thought, it's not a good test.

(4)

I doubt there has been sufficient time for sexual selection (which is probably weak to start with) to act on math/reading differences given the incredibly narrow time scale (widespread access to schooling is a recent, post Industrial revolution phenomenon). But this point is not relevant to the current discussion.

Oh, come off it!

"Miller explains the above sex differences with his "male-display, female-choice" logic. In short, sex differences in verbal abilities and language comprehension reveal that language evolved under sexual selection because men used language as a display (courtship) device, whereas women developed more acute language as an evaluation device."
http://philsci-archive.pitt.edu/5229/1/Mating-Inteligence-Virtues.pdf

The question would be whether there were regional differences in mathematical and verbal ability net of intelligence.

Whatever. (4) is irrelevant. You should at least acknowledge (1) in your paper. (2) & (3) would require a separate paper and review. I think your results are really weak here. But it's your name on the paper.

Add the gender index (see my SPSS file) & add a brief note about (1) and note that the results are tentative and in need of replication in your limitation section. Note possible methods by which this model could be tested in the future (e.g., compare race differences within country controlling for SES; conducting a "proper analysis" using samples discussed in reviews, or whatever) and I will approve.

"Miller explains the above sex differences with his "male-display, female-choice" logic. In short, sex differences in verbal abilities and language comprehension reveal that language evolved under sexual selection because men used language as a display (courtship) device, whereas women developed more acute language as an evaluation device."
http://philsci-archive.pitt.edu/5229/1/Mating-Inteligence-Virtues.pdf

G.Miller sometimes (more often than not) does not know what he's talking about. I read his book "The mating mind" full of totally unsupported claims and wild speculations that would make the "weak data" of my paper seem more solid than the great pyramids.
As far as I know his ideas on sexual selection of language (like many of his other ideas) are not backed by any solid evidence, and the Bracanovic paper is written by a philosopher (?) and anyway does not provide empirical data, but is just a philosophical discussion. If males are the selected sex as Miller seems to agree with me (the logic of male display=female choice) it's impossible that female language skills were selected for. According to that "logic" verbal ability could have been selected only in males, and the female advantage in reading comprehension actually falsifies Miller's claim. Adding the ad hoc explanation that women developed skills as an evaluation device is as close to a kids just so story as I have ever heard. However, this is not the place to discuss Miller's ideas, which are not relevant to this paper and for me the discussion about Miller's thesis ends here.

Whatever. (4) is irrelevant. You should at least acknowledge (1) in your paper. (2) & (3) would require a separate paper and review. I think your results are really weak here. But it's your name on the paper.

Add the gender index (see my SPSS file) & add a brief note about (1) and note that the results are tentative and in need of replication in your limitation section. Note possible methods by which this model could be tested in the future (e.g., compare race differences within country controlling for SES; conducting a "proper analysis" using samples discussed in reviews, or whatever) and I will approve.

I am working on modifying my paper trying to accomodate your requests and will upload a new manuscript soon.
As a preliminary report, the Gender Equality index has not affected my results much, it's actually made the correlation between CPS and sex differences a bit worse! And it's made the correlation between standard deviation and sex differences stronger! These conflicting effects are hard to interpret, also because the change is probably not significant. I guess it's due to the soft nature of the data, which makes it less objective in my opinion than GDP.