Sunday, October 16, 2016

Lynn's contrived work on IQ of India:-


Richard Lynn and Tatu Vanhenen presented several IQ studies on India's population numbering 13. As previously shown in my other blogs that his work on China and Africa are a result of manipulation of actual data available via "selective citation" or "cherry picking available data". There are many reports showing very low IQs in China and much higher IQs in Africa. Here I will show that his work on India is also a case of "selective citation" and far from being the truth.

As previously shown that present IQ data set on Africa is 78 IQ and in China it is 95 IQ. Though he did arrive at a decent number for China, 97.5; his numbers for Africa were way off and almost anyone (scientist and else) who has reviewed his work has come to the same conclusion.


Anyways, here are his samples on India :-




Based on this data, he calculates an IQ of 82 for India. First of all, his book isn't really a journal publication and rather an independent publication. But even if he had decided to publish this in some scientific journal, the scientific community would not accept it; not because they don't like the numbers and rather because the average IQ based on the table comes out to be 84.5 IQ. He is calculating "average" just by averaging the IQ numbers without really adjusting or weighing it for different sample size of the data (number of people tested) which indicates how carefully the book has been written.

Unlike how he makes a claim in his book to have done meta-analysis of IQ correlating for socio-economic and other factors (the standard ANOVA analysis) as if his work is a result of great award winning complex scientific calculations, in reality he hasn't even bothered calculating "average IQ" from a table properly which even a sixth grader can do.

As his book isn't a scientific publication, such work cannot be criticized at scientific level. And should not be taken as a result of the scientific community.

Anyways, here I will prove that many of his samples are contrived and will cite very high scoring samples on India.

Let me ask a very general question. How should the "average IQ" data-set of a nation look like. Well, it should have unbiased results of tests where people are selected randomly out of a given population irrespective of their caste, gender, religion, profession, etc. It doesn't really make sense to do IQ tests on say, Australian Aboriginals and show the data in average IQ for Australia.

This is what Lynn had done for India. Let me ask, which is the lowest IQ group within India ? Tribals, muslims, lower caste dalits ? Here are the IQ samples covered by Lynn in his book where the tests were only done on tribals, muslims without including the entire population which are clearly biased samples. And they number 3/13.

Mohanty and Babu 1983 (Average IQ: 79):-
As mentioned in the topic of the research article, the sample has been conducted on "Tribal population" in Orissa. Without any doubt, selection bias of "including only Tribal population" and excluding all "non-Tribals". Which leads to the fact that data is invalid.

Afzal 1988 (IQ of 79, IQ studies conducted in Urdu):
Abstract: "In order to study the effects of consanguinity on IQ, a survey was conducted among the Ansari Muslims of Bhagalpur residing in suburban and rural areas. Both outbred (N=390 from suburban areas and N=358 from rural areas) and inbred (N's=300 and 266, respectively) children aged 9 to 12 years from socioeconomically middle-class families were administered the Weschler Intelligence Scale for Children—Revised."







Even in many of his other samples, he is citing the results of the IQ tests of Asians living in India.

Majumdar, P. K., and Nundi, P. C. (1971)
http://psycnet.apa.org/psycinfo/1974-10149-001
Compared scores of 2 geographically separated Bengali-speaking groups of schoolchildren, grades IX-XI, on Raven's Standard Progressive Matrices. 2,836 Ss from Calcutta and 2,100 Ss from Agartala were tested. Ss from Calcutta were superior to those from Agartala by an average of 7 points.




Agrawal, N., Sinha, S. N., and Jensen, A. R. (1984)
http://link.springer.com/article/10.1007/BF01068128
Indian Muslim school boys, ages 13 to 15 years, whose parents are first cousins, were compared with classmates whose parents are genitically unrelated on the Raven Standard Progressive Matrices, a nonverbal test of intelligence.

It is well known that people in North Eastern part of India are genetically same as South East Asians and not Indians or South Asians. 

So the question arises, why is he using the results for "Muslims, Tribals, Asian(non-Indian) minorities" in his book.

The obvious answer is "the low IQ results". North East Indians (which are the same as South East Asians), 3 Muslim samples and Orissa Tribal samples are the only samples on India showing IQ less than 80. That's 4/13 samples just on the minority population. 

Even in the Majumdar and Nundi sample, people in India scored 7 IQ points above North East Indians. 80 for North East Indians and 87 for Bengal.

Now the question arises, are there any high scoring IQ samples on India. Yes indeed there are many samples showing 100+ and even more showing 90+ IQ via random testing on India.

SA Khan 2015:-
http://www.jcdr.net/article_fulltext.asp?issn=0973-709x&year=2015&month=November&volume=9&issue=11&page=ZC10&id=6726

Average IQ of school kids in villages near Lucknow in non-dental fluorosis regions was 110 and 100 overall (normal fluoride region and severe fluoride regions combined).

IQ of Ahmedabad (sample size: 190):-
Source: http://www.fluoridealert.org/wp-content/uploads/trivedi-2007.pdf


Average IQ of villages near Ahmedabad was found to be 104 and 97.  





http://hrcak.srce.hr/file/44025

IQ of Muslims in Aligarh city, Uttar Pradesh (Average IQ of 107)

Source: “Badaruddoza. Inbreeding effects on metrical phenotypes among North Indian Children. Collegicum Antropologicum 28(Suppl. 2): 311-318. (ISSN No. 0350-6134.”)


IQ in Maharashtra (Pune):-

http://www.indianpediatrics.net/feb2004/feb-121-128.htm

86.1 for low birth weight, 92.1 for medium birth weight and 97.2 for normal birth weight.

https://www.indianpediatrics.net/july1999/july-669-676.htm


IQ of control group was 101.38. And LBW was 94.

https://www.researchgate.net/publication/297722925_Dental_fluorosis_and_urinary_fluoride_concentration_as_a_reflection_of_fluoride_exposure_and_its_impact_on_IQ_level_and_BMI_of_children_of_Laxmisagar_Simlapal_Block_of_Bankura_District_WB_India?_sg=OxEw86Y9iI6rPYqTlpXNxf3sHxsD4oLarEkZXw_AKdiC1SmLw9IzIsEDusa540LU


IQ in Villages in Laxmisagar Village was 108.3 in normal regions and 85–92 in fluoride contaminated areas.


IQ of Chennai, Tamil Nadu:-

Study 1:-

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3018494/

IQ of 107 as a community sample. 4 zones in Chennai and 12 schools were selected randomly. Sample size of 717. 606 children belonged to families with less than 6500 INR monthly income. And 130 had illiterate mothers. If anything, the selection bias towards socio-economic status is downward.

Study 2:-

http://www.tandfonline.com/doi/abs/10.1179/oeh.2005.11.2.138


IQ of children was found to be 102. And 95 for those having high lead in blood level.

IQ in Chandigarh, Punjab:-

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3767262/


Average IQ is 99. 5 private schools and 5 government schools were randomly selected out of 37 schools. And sample size is 2400.

IQ in 2 rural and 2 urban areas of Punjab:-

http://www.ijres.org/papers/Volume%204/v4-i7/Version-2/G4724854.pdf

IQ in 2 rural areas is 96.2, 90.4. And 2 urban areas is 102, 106. IQ grades are available (0 to 10th percentile, etc.) and online statistical tool is required to read.


Rural areas are in Dera Bassi and Tehsil Kharar. Urban areas is in Fatehgarh Saheb. And Tehsil Kharar.

There are many articles on India showing IQ higher than 90. Even the Lynn's data shows IQ of 86 on India. One can easily conclude that long term IQ potential of India (just by removing mal-nourishment) is 97, same as Europe. And Lynn also estimated the same IQ for Indians in the long run.

Now, lets look at the TIMSS and PISA scores of India (which are much better way of sampling, but do not measure the genes):-

This is an article of India’s TIMSS performance.
Poorest states (not known for IQ) were selected in TIMSS. I will compare India to European nations and South East Asia.
New Zealand: 494, Orrisa(India): 404, Rajasthan(India): 382, Philippines: 378
One standard deviation in TIMSS score is 100 points. Even the poorest Indian states are performing at 85 IQ ahead of Philippines, Saudi Arabia. Overall Indian IQ cannot be less than 85 at any cost as even poorest regions in India are showing 85 IQ on TIMSS.



PISA 2009 reports on India show an IQ of 74-77 in Himachal Pradesh and Tamil Nadu, two high scoring states.

PISA sample of India is way off from other data-sets available on India's population.

As India has a school enrollment rate of 96%, TIMSS and PISA for India is the real indicator and not cherry picked IQ samples shown by Richard Lynn. 

Though it was almost conclusive from TIMSS and IQ reports on India that long term IQ of Indians is at par with Europe (97), PISA 2009 results have opened up different possibilities for Indian average IQ or educational systems in India. Detailed PISA results on India are unavailable and it is hard to predict whether Indian PISA takers lacked "cognitive ability" or "mathematical equations" in PISA.  

As the results are contradicting, nothing can be said as of now. And a few more rounds of PISA and TIMSS will make things clear.

On the contrary, IQ and test scores on Indians living in Australia, UK, USA, Singapore, Malaysia, Hong Kong, Canada, UAE, South Africa, Trinidad and Tobago, Mauritius indicate IQ of 97-105 among average Indian.

As IQ is a better measure of genes than PISA and TIMSS, it is almost certain that current IQ of Indians is 89 and will shoot up to 97 on removal of mal-nourishment.

5 comments:

  1. I just want to point that out that the statement "North-East Indians" are the same as "South East Asians" is inaccurate, South East Asia has historically received a lot of migrations and invasions from the India-South Asia region historically as well as colonization's. Much of the culture and language of South East Asia is entirely based on Indian culture from their script and alphabet to their language, and there is a lot of Indian India origin genes in South East Asia.

    South East Asia region is diverse in itself with people of varying phenotypes and genes, and it's not just filled with people of the "Mongoloid" phenotype either. So this statement that people of North-East India are exactly similar to South East Asians is inaccurate, as they are both Indic (India) types as well as "Mongoloid" types in that region.

    However there is evidence that the Mongoloid inhabitants of modern day NE India are infact recent transplants and migrants brought in by the British from modern day Southern China. Infact colonial European powers like the British deliberately and often imported Southern Chinese immigrants into places like Malaysia and the Carribbean and even Africa where they were routinely used as a middle men buffer class to keep the indigenous native populations down or use them as a buffer shield sand which and the British were not the only European colonial powers that did this (the Dutch also imported large numbers of Southern Chinese origin migrants to places like Indonesia etc etc).

    So given that the British controlled places like Burma and built up places like Hong Kong, there is a lot of evidence that has been uncovered by independent researchers that the modern NE Mongoloid Indians are infact recent Southern Chinese transplants and migrants brought into the region, and this is unfeasible considering the region is practically right below and near the Southern China region geographically.

    So the actual IQ's of these tribal NE Indians would infact be the real IQ of the Mongoloid inhabitants of Southern China without any sort of bias thrown in and with the added factors of low education and possible low nutrition.

    Just wanted to point that out......

    ReplyDelete
    Replies
    1. and this is completely feasible and completely possible considering the region is practically right below and near the Southern China region geographically*******

      Delete
  2. Sorry man but IQ doesn't work like this. It's a relative scale not an absolute scale, the same number can mean something else for different populations. It's not like tests you got in school where you got certain number of points out of some maximum amount.

    To make different IQ tests comparable with each other you have to standardize them first. Lynn standardized them in such a way that British IQ was set to 100 and all estimates were recalculated in relation to that.

    The papers you are posting are standardized on Indian population where 100 IQ is Indian mean. You can not conclude from this that Indian IQ is the same as in Europe.

    ReplyDelete
    Replies
    1. You can for example standardize your IQ scale on Somali population and then average Somali IQ would be 100. Then someone would make study where they found that malnourished Somalis had IQ of 94 while well-fed Somalis had IQ of 107.
      Then you would come along and say "see? After we remove malnourishement in a country Somali IQ will be 107. Higher than in Hong Kong!"

      This is obviously silly.

      Delete
    2. lies. mensa did an iq test on 6000 poor under privileged children around delhi with their international standard iq test. least iq was 85 and 12 children made it above 145 with median iq of 100.

      Delete