Hello There, Guest!  
 Previous 1 2 3 4 5 9 Next   

[ODP] The Scandinavian WAIS IV Matrices as a Test of Dutton, te Nijenhuis and Rovaine

#21
(2014-Oct-29, 16:52:04)Emil Wrote:
(2014-Oct-29, 07:43:59)Philbrick Bastinado Wrote: Some grammatical errors:
"Although Finland has a higher fluid g than Scandinavia but its CPS native score is only slightly higher than the rest of north­western Europe, at 526 points, giving a Greenwich fluid g estimate of 101.4."
"This is possible, however a meta­analysis..." This should read, "possible; however, a meta-analysis...".
"the presence of significant Northeast Asian admixture in the Finnish population" Citation, please.

I prefer the term "fluid intelligence" (or, even better "fluid skills") to "fluid g". This is quibbling, however. (Also, g should be italicized.)


E.g. http://www.eupedia.com/europe/autosomal_...ecad.shtml

[Image: East-Asian-admixture.gif]


OK, but the citation should be in the paper.
 Reply
#22
(2014-Oct-29, 01:55:26)Meng Hu Wrote:
Quote:We used 96 as the SD of CPS scores as done by Piffer and Lynn (2014). 96 is the OCSE average SD. It is better than using country SDs because using individual countries SDs inflates the IQ score of countries with lower SD.

I would like the last sentence to be more elaborated.


Below I report the justification for the use of the average OECD's SD, and why using individual country SDs overestimates the IQ of those with lower SD.
Let countries X and Y have the same PISA score which is 50 points higher (550= 500+50) than the mean OECD score . However, country X has SD 80 and country Y has SD 110. Thus, country X's score would be(50/80=) 0.625 SDs and country Y's score would be(50/110=) 0.454 SDs higher than the OECD average, despite having the same PISA score. Thus, a PISA score of 550 would correspond to two IQS of 109.4 and 106.8. This result is clearly absurd.
 Reply
#23
Thanks. That should be added in the article.
 Reply
#24
Thank you, reviewers, for your ideas, time and patience. I have now made all of the most recently requested changes to the article.
 Reply
#25
(2014-Nov-03, 10:36:52)Barleymow Wrote: Thank you, reviewers, for your ideas, time and patience. I have now made all of the most recently requested changes to the article.


Do you refer to this version ?
https://docs.google.com/document/d/1QqVs...AhfRE/edit#

You should have said it in your post, otherwise it's difficult to guess. Since that file is constantly updated, you should probably add this in the initial post :
http://openpsych.net/forum/showthread.php?tid=140

By the way, I think you must be careful with the file. For example, I see that I'm able to edit the article. You should restrict this function to co-authors only. An accident can easily happen, otherwise.

Regardless, I see the changes you have made. I think you should talk about the influence that politics has on educational systems (not just g, if we accept the very odd idea that politics cause IQ/g). But that's really the least of my problem. I can live without that.

My problem has to do with your conclusion. You say finnish people have higher Gf, given CPS PISA scores, even though WAIS-IV matrices show finnish people have lower Gf, because PISA data has larger N. Ok, but if you average the two, you will see that the difference is not obvious anymore. You didn't talk about the differences in tests. Gf is measured with only two tests. I don't know the properties of WAIS matrices (perhaps someone can show me a sample of what the items in that subtest may look like) but if it's different from CPS PISA, perhaps it's that difference that may explain why you get conflicting results. Also, I still don't know which of these two tests are the best approximate of Gf.

Do you have other evidence (direct or indirect) that may support the idea that finnish people have higher IQ and/or Gf, other than this one ?

And finally :

Quote:This is possible, however a meta-analysis by Kirkegaard (2014) of all the PISA results showed only weak evidence of conscientiousness explaining variance that was not explainable by measured IQs, and this was only for reading (standardized β = .17, p = .03) not for the CPS test which we used here (standardized β = -.03).

Given the reference, it's just a blog post. However, it's not the central topic of the article, so I won't cry. But I clearly prefer to see a peer-reviewed paper when you cite an analysis. If you cite an article, say, a review, or a commentary, it's fine to cite blog posts. I don't know if others will agree with me, but that's what I think.
 Reply
#26
CPS and the WAIS IV matrices are highly comparable,which is why we used them.
 Reply
#27
(2014-Nov-03, 21:46:32)Meng Hu Wrote: My problem has to do with your conclusion. You say finnish people have higher Gf, given CPS PISA scores, even though WAIS-IV matrices show finnish people have lower Gf, because PISA data has larger N. Ok, but if you average the two, you will see that the difference is not obvious anymore. You didn't talk about the differences in tests. Gf is measured with only two tests. I don't know the properties of WAIS matrices (perhaps someone can show me a sample of what the items in that subtest may look like) but if it's different from CPS PISA, perhaps it's that difference that may explain why you get conflicting results. Also, I still don't know which of these two tests are the best approximate of Gf.


I agree with MH that there is not enough evidence that Finns have higher gf. There is much stronger evidence from PISA math, reading and science that they have higher scholastic aptitude or crystallized g, but this points towards an explanation in terms of better schools or higher C, which are likely to pump up scholastic intelligence much more so than fluid g.
Swedes living in Finland score lower in PISA CPS (there is data on this in the PISA report) and this may argue against a cultural explanation. The authors should say this.
However, a note of caution is necessary because the evidence for higher fluid g is not strong and the discrepancy between tests of scholastic aptitude and fluid intelligence may require a cultural explanation.
I'll approve publication provided these issues are dealt with in the discussion.
 Reply
#28
A blog post is no different than citing other grey literature, e.g. unpublished dissertations. These are frequently cited in papers, especially meta-analysis. Recall that peer-review is a relatively new entry to science, and was not widely used before recent decades. E.g. Einstein's famous papers were not reviewed. I did the meta-analysis on my blog because I already have 3 papers in review (the maximum). I cannot submit more stuff yet, so it will have to wait until the other papers are published/retracted.

-

The sample sizes of PISA CPS are not that much larger than the WAIS matrices. Taking a weighted mean would give Finns a small advantage, which may not be significant. This is the proper procedure.

-

Another thing when comparing is to adjust for the immigrants or not. When talking about the high gf, g or IQ of Finns, sometimes one means the country as a whole, sometimes the genetic Finns. Different explanations apply. Genetic explanations concern only the Natives (excluding Finland Swedes), while educational system explanations cover immigrants too.

Here's the CPS data for whole population and Native:

Code:
ID    CPS12Mean PISA12CPS_Native
ARE       411              376
AUS       523              524
AUT       506              516
BEL       508              522
BGR       402              405
BRA       428              431
CAN       526              532
CHL       448              448
COL       399              400
CYP       445              447
CZE       509              510
DEU       509              523
DNK       497              505
ENG       517              523
ESP       477              482
EST       515              519
FIN       523              526
FRA       511              523
HKG       540              545
HRV       466              467
HUN       459              459
IRL       498              501
ISR       454              452
ITA       510              514
JPN       552              553
KOR       561              562
MAC       540              538
MNE       407              406
MYS       422              424
NLD       511              520
NOR       503              510
POL       481              482
PRT       494              498
RUS       489              490
SGP       562              561
SHA       536              538
SRB       473              474
SVK       483              485
SVN       476              481
SWE       491              501
TUR       454              455
TWN       534              535
URY       403              405
USA       508              512


The four rows of interest are:

Code:
ID    CPS12Mean PISA12CPS_Native
DNK       497              505
FIN       523              526
NOR       503              510
SWE       491              501


We see that Denmark loses 8 points by immigration, Norway 7, Sweden 10, but Finland only 3. There are in fact fewer immigrants in Finland than in the other three. Sweden has the most and also lost the most points. Norway has more than Denmark, but Denmark lost 1 point more. Perhaps due to exclusion rate differences or because their immigrants are different. But the correlation between points lost and immigrant% is strong.

So we see that depending on whether we are comparing Natives or not, the difference changes. It is not that large when comparing only Natives, about 505 vs. 526.
 Reply
#29
I agree with Piffer that schooling in Finland could have improved Gc more than Gf, which may explain why you have evidence of stronger g but weak evidence of stronger Gf. But if really there is no difference between Finland and Scandinavian countries in Gf, that also means that the higher g found among finnish people is merely due to higher Gc. This suggests two things; that the higher IQ is not general, but domain-specific. Second, that if education is responsible for the higher Gc but not Gf, then schooling effect in this example shows no g gains because it's domain-specific.

Emil, if I'm not mistaken, dissertations or theses are being approved by professionals. Like peer-reviewed papers.
 Reply
#30
They are graded and they can fail their defense (rarely happens), but they are still cite-able and not approved in the sense that peer review for a journal is. Furthermore, usually the profs grading it are the advisers, which means they are biased at their review.
 Reply
 Previous 1 2 3 4 5 9 Next   
 
 
Forum Jump:

Users browsing this thread: 1 Guest(s)