Given the current version, I don't have much more to say. Although some points were not addressed appropriately, I don't think it's too big of an issue. There is only one thing I require before accepting: update the supplementary files, along with any missing files mentioned prior. The code is still the same as before (uploaded at 2025-01-06). Remember this is required for all OP submissions. 

Did you update the supplementary files (especially the code)? I still don't see any update in this regard.

Given the current version, I don't have much more to say. Although some points were not addressed appropriately, I don't think it's too big of an issue. There is only one thing I require before accepting: update the supplementary files, along with any missing files mentioned prior. The code is still the same as before (uploaded at 2025-01-06). Remember this is required for all OP submissions. 

Did you update the supplementary files (especially the code)? I still don't see any update in this regard.

I am a bit late, but below is a copy of a review I received recently (19th May) for this paper. I can only copy the content as uploading the screenshot still doesn't work properly.

I encourage the authors to provide some response to all of the comments. 

(1) Fix typos, e.g.: "and they found that 5 out of 7 found a small female advantage in general intelligence" or "13 year olds

Fixed.

(2) Did you consider searching on PubMed or Web of Science (in addition to Yandex...)?

Not at the time. 

(3) Why are you describing datasets that you ended up not using (e.g. GSS)? A simple mention will suffice.

I suppose, though removing the text at this point seems like a waste.. 

(4) The PPVY only seems to capture verbal intelligence. Why did you include this but not WORDSUM? Consider dropping it.

The PPVT is more exhaustive and includes more questions. I don't exclude tests from the standard meta-analysis for only testing one ability. 

(5) You say that the issue of latent differences was ignored, and that only 2.8% of studies examined this. For those unfamiliar with this literature (such as myself), can you elaborate on how this technique differs from looking at sex differences in the latent general factors, or from the method of correlated vectors? How is it possible, as Pezzuti & Orsini (2016) claim, that "the observed difference in intelligence [in favor of men] is of roughly the same magnitude as the latent difference [in favor of women]"? Can you comment on this further?

>the observed difference in intelligence [in favor of men] is of roughly the same magnitude as the latent difference [in favor of women]"?

Neither I or they claimed that there was an observed difference in intelligence in favour of men. I commented on the latent/observed difference in the discussion section, and added a section in the methods section where I define what a "latent difference" is.

(6) I don’t understand what you are correlating with what on Table 3. Please explain more clearly. For example, is .83 the correlation between the sex difference in reading and the sex difference in scientific comprehension across nations?

Vector 1 in sex differences in sci lit by country, vector 2 is sex diffs in reading by country, vector 3 is sex diffs in math lit by country. Table 3 is the correlation matrix.

Good paper. I thought it very mind-provoking when I read Hanania's blogpost about this. I would like to see more on why the method of correlated vectors produces null results. The same method has of course been used to support race differences in g, with predictable results.

Also, I just noticed that you didn't skip a row regarding the title of Figure 4. Please do so.

I looked at the latest version, just out of curiosity. You have a new typo.

This data was excluded from the final analysis over concerns about the reliability and validity of the WORDSUM test, which has only 10 items. .

There's an extra dot at the end.

It has been removed.

the sex difference in intelligence would be relatively small in magnitude and impact assuming it exists

The old version didn't have "and impact" and I think it was better. I think it's confusing to have magnitude and impact, as if you meant those are two different concepts (in what way, then?). 

It has been removed. And they do differ: one is x, and the other is x * b.

I also realize now about a detail I didn't notice earlier. In the figures title, you have "Fig 1" "Fig 2" "Fig 3" "Fig 4" and then for some reason it became "Fig. 5" etc. Try to be consistent. Also, why "Fig 1" ... "Fig. 5" and so forth are in bold, but Tables 1-3, including Table A1-A2 are not in bold? Again, be consistent.

Fixed.

After "9. References" please jump a line.

Fixed.

I probably should have mentioned this before (months ago in fact) but your new code is not working anymore, because you didn't load the packages. In the old code, I remember the code worked, partially (some functions didn't because of some missing files).

The code still does not work properly. There are missing objects, such as calculate_average, iqresults, zxc2, etc. And some descriptive statistics are still not displaying any results due to having numerous warning messages. Can you review the code carefully, and make sure it works? Make sure you also check the results displayed in the paper, because, for some results I couldn't find where you got the numbers. One of which was Table A2. Perhaps you updated the code and forgot to update the numbers in Table A2, or something else.

OSF updated. I suppose I could share the analysis in A2, but the dataset is supposed to be private. 

Another difficulty with reading the code is how you name the objects (similar to how you name the datafiles and your OSF project), such as zxc2 (what does that even mean?), making it very hard to understand what is the relevant analysis. At this point, I won't ask you to rename all of these obscure objects' names, but for your future papers, you should avoid this.

I am still waiting for the code to be updated. Like I said before, the code fails in multiple parts. And it's not just because of iqresults csv file lacking. Many objects just don't run. Can you fix your code? There are still multiple warnings in your code. That's why I asked you to check very carefully every single output you get.

Does an Rpubs work?

https://rpubs.com/sebjenseb/1372678

Results in Table A2 still don't match the numbers I get when running your code, so how can I trust your code then? In your OSF, you only added iqresults csv, and it has nothing to do with Table A2 results.

I redid the regressions in the Rpubs and found the same results.