Old plebbit post I'm reposting here

You’ve probably heard it before: Sierra Leone’s IQ is 91

I grew too tired of this number being thrown by lazy detractors around as if it were legitimate, so I decided to make a post about it. First, I've only ever see the number of 91 for Sierra Leone to come from one place - a website "Brainstats.com". You can see Sierra Leone at 91 tied in the list for #16, which places it above some Slavic nations like Albania, Croatia, and Serbia. Opponents obviously love this reference because it shows a sub-Saharan African country listed as higher than some European countries. Of course, proponents have no idea where 91 comes from (even if they pretend) and Sierra Leone is just one small country in a very low-IQ region, so it would at best be an exception to a general rule. Or even just how an average IQ of 91 seems dubious, given that Sierra Leone has a GDP per capita of $1800*_per_capita) – 181st out of 198 countries. None of that matters – Slavs < Kangz, Balkans eternally BTFO

Intuitively, it doesn’t make sense that a small sub-Saharan African country of Sierra Leone would have such a high average IQ – because it is an extreme outlier. Its neighbors, Liberia and Guinea, both have IQs of 67 respectively in the very same list (with the rest of Sub-Saharan Africa). A number of 91 for Sierra Leone should be very unexpected. It could be argued that tribes in Sierra Leone are just “smarter” then their neighbors or something – but we know Africa's borders were largely drawn arbitrarily by European powers, and don't reflect actual tribal territories. So, how did the little Sierra Leone get so smart - almost twenty points above its neighboring countries (but stay so poor)?

The short and obvious answer of course is "it didn’t" - 91 is probably some kind of measurement error or otherwise from a very elite sample in this country. So, what sources back up BrainStats.com’s number of 91? The site lists three:

IQ and the Wealth of Nations (2006)

IQ and Global Inequality (2002)

Studies of national cognitive ability (Wikiwand)

Both of the first two works were completed by the same authors – Richard Lynn, and Tatu Vanhanen. I have never, seen either work in completeness on the internet for free before despite searching for them many times. Note also the author of this site has the dates wrong. IQ and the Wealth of Nations was released first in 2002, and IQ and Global Inequality in 2006, which included some responses to critics on the first work. The final link is to another website, and we’ll look at it later.

So what do Lynn and Vanhanen report the IQ of Sierra Leone to be? Well, neither Wikipedia article has the listing, but the closest thing I ever found to a real copy of these is this PDF. Sierra Leone’s IQ is listed as 64, in both 2002 and 2006. This matches the two studies listed for Sierra Leone in Richard Lynn’s Race Differences in Intelligence, published in 2004 (Table 4.1). So, whether that PDF is legitimate or not, Lynn obviously believed the IQ of Sierra Leone to be 64 ion 2004, not 91, which matches its other Sub-Saharan neighbors, and it is almost definitely 64 in the 2002/2006 works cited by BrainStats. So, if the number can’t come from Lynn and Vanhanen, where does it come from?

Well, there have many criticisms of Lynn and his work over the years (he is a racist, after all). The main critiques come from Jelte Wicherts and his team of Dutch researchöörs. The argument is mostly that Lynn uses samples from diseased, starving, illiterate, not smart enough, or otherwise disqualifiable African populations, and that this understates the true IQ of Sub-Saharan Africa. AltHype talks about this and gives a good overview:

Jelte Wicherts criticized Lynn’s selection of studies. He argued that Lynn included studies with developmentally disabled, with people who didn’t understand the test, with people who have HIV, malaria and parasites.

Lynn’s response was basically that those things represent black Africa, and to exclude those studies is to, in essence, exclude the problems of Africa. And Lynn and Wicherts went back and forth on that.

Wicherts estimates the median IQ of Sub-Saharan Africa to be 80. And he does this by excluding studies that had the problems he referred to above.

As an aside - accepting a Sub-Saharan African IQ of 80 kills the egalitarian. If pure-blood Africans living in Africa with some of the lowest standards of living in the world score 80, and Black Americans (that are 20-25% white on average) with average incomes of $39k/yr score 85, this would imply that the differences in IQ are almost 100% genetic and the environment is worth almost nothing. Also, so you know where AltHype stands on this:

Lynn’s number: 70.01

Average of Lynn and international test scores: 72

International Test Scores: 74

Average of all 3 numbers: 74.67

Average of Wicherts and international test scores: 77

Wicherts’ estimate: 80

My opinion is 72.

So bringing this back into context - a number of 91 Sierra Leone cannot possibly come from Lynn and Vanhanen, and is far-and-away above even the most generous estimates for Sub-Saharan Africa on average by Team Wicherts. So where does the number of 91 come from? Clicking on “Sierra Leone” Brainstat’s site brings us to this page which is equally unhelpful. It leaves us with only one cited source left, this website “Wikiwand.com: Nations and Intelligence”

This site gives a brief overview of Lynn and Vanhanen’s work, the criticisms of Wicherts and others, and some data from world test scores. It also provides some other data on correlates with IQ and explanations for the differences (including genetics). Overall, it’s a pretty balanced and brief overview of all the relevant literature. However: There is no listing of IQ scores by nation here at all - or even a mention of Sierra Leone.

So.. where does BrainStats.com get 91? It doesn’t come from the sources listed, at least not in an obvious way. I had a hunch, though – Wicherts. This researcher/team are very dedicated in their opposition to Lynn/Venhanens' work and their methodologies, and seem highly-motivated to disprove them.

From A Systematic Literature Review of the Average IQ of Sub-Saharan Africans by Wicherts, Dolan, and van der Maas (2010) we get our answer. There are two studies for Sierra Leone listed here that Lynn and Vanhanen don’t use. The first is a study of 400 Students from Freetown, IQ = 77.8. The second is 202 Children, IQ = 93 (Ohuche and Ohuche, 1973) (Table 2) The authors actually discuss Sierra Leone in more detail in their = paper (page 12). Here they discuss their dismissal of the two studies that reported 64 for Sierra Leone:

Ferron (1965) refers to a study by Dunstan (1961, cited in Ferron, 1965) on a sample of “over 400” form I students in Freetown, Sierra Leone. These students took one of the Moray House Tests (no. 44) and Ferron indicated that their mental-age IQ was around 77.8 (i.e., they had a chronological age of 13.5 and a mental-age of 10.5). Ferron indicated that Dunstan reported problems during test administration and a “language handicap”. These data do not meet our inclusion criteria

So basically, Wicherts and others make up criteria (section 3.2) that conveniently disqualifies all other tests except the highest-scoring test for Sierra Leone, a test of 200 School children where their IQ was 91.3 (Table 5). So Wicherts et al. do more rigorous research, Sierra Leone’s IQ is 91, not 64 like Lynn and Vanhanen reported. Case closed?

Nope. Even if most researchers agreed with Wicherts et al.’s exclusion criteria (likely not), 91 is easily dismissed as an inaccurate estimate for adults Sierra Leone based on our knowledge of general sub-Saharan African IQ, Black American IQ, and the Wilson Effect (increasing heritability with age. For instance, it is well-known that the racial gaps in IQ in the US start off smaller in childhood and increase with age (Rushton and Jensen, 2005). Worse yet, the test given to children was the “Draw a Man (DAM)” test. This barely qualifies as an IQ test, and doesn’t correlate very well with other tests, like the Wechsler (r=0.27). So this test probably isn’t even a good indicator of childhood intelligence, period.

Lynn and Miesenberg respond to the criticisms from Wicherts et al. in The average IQ of sub-Saharan Africans: Comments on Wicherts, Dolan, and van der Maas (2010)

WDCM [sic, Wicherts, Dolan, Carlson, and van der Maas 2010, a related paper] propose nine inclusion criteria for the acceptability of studies of the IQ in sub-Saharan Africa, but these do not include the crucial criterion that the African samples should be representative of the population. This is a strict criterion because there are no perfectly representative samples from sub-Saharan Africa. We therefore have to make judgments on which studies are sufficiently representative to use. We do not follow WDM in rejecting studies in which (1) “Test administration should not be described as problematic” because this means that the samples lacked the cognitive ability to understand the instructions and/or the test was too difficult; (2) sub-Saharan Africans are compared with matched whites, because in some studies this comparison is more appropriate

On Sierra Leone:

Ferron (1965). This study reports IQs for seven samples of children in Nigeria and Sierra Leone that included three grammar school samples (IQs 91, 95, and 81), two samples taking the entrance exam to grammar schools (IQs given as 80+ and 70+), and two samples from two primary schools. WDM average the seven samples to give an IQ of “around 77”. Comment: The first five samples should be excluded because grammar school students are selected for higher IQs and the imprecise IQs given for those taking the entrance exams to grammar schools are unusable. Ferron's two samples from primary schools are acceptable. The IQs of these were 74 and 66, averaged to 70.

Ohuche and Ohuche (1973). WDM [sic, the aforementioned Weicherts, Dolan, and van der Mass 2010] calculate an IQ of 91.3 for this sample of children at an experimental school in Sierra Leone. Comment: The sample is unrepresentative; the ages of the children are unknown; all children repeating the year (i.e. those with low IQs) were excluded; the IQ of 5–6 year olds= 69.5; the IQ of 7–12 year olds= 94.2, a discrepancy indicating serious problems with the data; there was no correlation between IQs and tests of English, math and social science in grades 4–7, showing IQs have no validity for these ages. The study is so unsatisfactory it has to be rejected.

Oof. So why did BrainStats.com choose an IQ of 91 for Sierra Leone based on a questionable stickman-drawing test done by school children? Why would Wicherts et al. even choose to include it? I can’t say. Regardless, the BrainStats.com citation is wrong, this single study is a huge outlier, it probably doesn’t give an accurate measure of intelligence, and is definitely innapropriate to use as an estimate for the overall adult population of Sierra Leone. I can’t speak to the motivations of the authors of this site – but they seem to really want you to take their Free IQ Test!

Obviously a website called BrainStats.com is not a serious reference. If you are going to make the outlandish claim that a small, sub-Sharan African country has an IQ >17 points above its neighbors, you should probably know where that number comes from. We still have an unanswered question though: What is the average IQ for Sierra Leone? Let’s be generous to all parties: We’ll include all studies and stick man drawings, no exclusions or adjustments:

Berry (1966): 64

Binnie-Dawson (1984): 64

Ferron (1965) (Sierra Leone and Nigeria, 5 studies) “About 77” (Wicherts)

Ohuche and Ohuche (1973): 91

Average for Sierra Leone: 74 (absurdly generous)

TL;DR: Brainstats.com cherry picked (intentionally, or not) a single study from Wichert’s et al. (2010) that reported an IQ of 91 for schoolchildren based on the “Draw a Man” test - one that isn’t even a good measure of intelligence (1)(2), and is wildly inappropriate as an estimate for average adult IQ in Sierra Leone. The real average IQ of Sierra Leone is generously estimated to be 74, and definitely nowhere near 91.

Take your free IQ test now!