ArticlePDF Available

Solid numbers, missed opportunities: Review of The intelligence of nations.

  • Ulster Institute for Social Research

Abstract and Figures

(there is no abstract)
Content may be subject to copyright.
Solid numbers, missed opportunities: review of The Intelligence of Nations
by Richard Lynn and David Becker
Emil O. W. Kirkegaard, independent researcher, Denmark
Invited book review to Evolutionary Behavioral Sciences
The Intelligence of Nations
is the latest iteration of Richard Lynn’s long running
compilation of data on the intelligence of countries. The book has four chapters, the first
of which briefly (6 pages) introduces the reader to the question of why some countries
are richer than others (Smith, 1776). This chapter serves to summarize the three
previous books by Lynn and Vanhanen on the same topic (Lynn & Vanhanen, 2002,
2006, 2012). In his summary, however, Lynn neglects to mention his own 1978 book
chapter (Lynn, 1978) where he first summarized test results for ethnic groups and
countries. In that paper, Lynn did not produce a table of estimates, so I have constructed
one based on his reported studies; the corresponding values from the present book are
reported alongside in Table 1 for comparison.
Table 1 - Estimates of national IQs of select countries, 1978 and 2019
1978 values are based on (Lynn, 1978), using medians to combine studies when
needed. 2019 values are based on Lynn and Becker’s book.
1978 IQ
1978 notes
2019 IQ
United Kingdom
United States
European descent
New Zealand
European descent
East Germany
elite samples
elite samples
South Africa
African descent
Bandung, elite
The 1978 estimates were often from unrepresentative elite samples, and there were no
adjustments made for the Flynn effect, which was not brought to significant attention until
6 years later by James Flynn (Flynn, 1984). Still, the correlation between the 1978 and
2019 values is r
= .81 (n = 23). This is quite strong, especially given that the early
estimates were based on approximately 30 studies, whereas the latest dataset is based
on 667 studies (version 1.3.1, the one found in the book).
Considering the stability of results over time, the reader might wonder: What’s new in this
book, as compared to the previous works? First and most foremost, Tatu Vanhanen
passed away in 2015, and so Lynn recruited the much younger (born 1983) German
intelligence researcher David Becker to assist him. The previous editions were
sometimes criticized for opaque methods and faulty calculations (e.g., Barnett &
Williams, 2004). To address these criticisms and to remove errors in the calculations, the
authors set out to re-run all calculations for the present book. They also set out to bring
their methodology in line with emerging principles for open science, including open data,
methods, and code (Hesse, 2018; Lindsay, 2017; Paxton & Tullett, 2019). The authors
therefore sought to obtain a copy of every paper that had been used. However, many of
these studies have not yet been obtained (514 accounted for, 70%, with 216 still
missing), and so were excluded from the present edition of the book. This is not so
surprising, given that many of the original articles were published, over the last century,
in extremely obscure, sometimes non-English outlets that no longer exist.
The main take-away from the new set of results is that the values estimated in earlier
works hold up well under increased methodological rigour and recalculation by a new
researcher (Becker did the calculations, Lynn provided copies of the sources). It’s worth
saying something about the new methods employed, which is the topic of the second
chapter. The authors illustrate their methodology by inviting the reader to consider data
for two fictional countries. They provide various characteristics about the fictive studies,
tests used, test administration year, test standardization year, representativeness,
sample size, age span, and so on. Then they introduce the reader to the various ways of
adjusting for distortive effects (including the Flynn effect which differs both by test and by
region), and the various ways of weighting the results by study quality. The process ends
up being somewhat similar to the one used for Cochrane systematic reviews, though the
authors do not seem to take inspiration from any guidelines for meta-analyses (e.g.
PRISMA, Swartz, 2011). Next, the authors discuss methods for computing national IQs
from international scholastic achievement/assessment tests (PISA, TIMSS, etc.), as was
also done by Heiner Rindermann earlier (Rindermann, 2007, 2018). After the process is
explained, tables are presented with the new IQ estimates. However, it’s unlikely that
anyone will be using these tables specifically because, as the authors explain, the
national IQs are now more frequently updated and are available on David Becker’s
website for download in machine-friendly format ( The current
version is 1.3.2, which contains several small error fixes compared to the values in the
book, and adds a few new studies. This approach brings the database of national IQs up
to par with other widely used datasets, such as the Maddison Project
( which provides national
economic measures, and Clio-Infra ( which provides historical data of
interest to economic historians.
Of particular interest to the study of group differences in intelligence are the comparisons
with the previous data sets, since so much research has been conducted on these. It has
been frequently claimed that Lynn’s national IQs were biased in favor of European
nations and against African ones (Rindermann, 2013a; Wicherts, Dolan, & van der Maas,
2010). Such bias does not need to result from deliberate actions, but can occur as part of
routine scientific work, which involves making make judgment calls where preconceived
notions might have an effect. In fact, the new IQs correlate at r = .85 with the Lynn and
Vanhannen’s (2012) shown in Figure 1. Since there is a huge overlap in sources, one
might wonder why this value isn’t higher. The reason seems to lie mostly with the use of
expanded norm ranges. Lynn did not previously rely on extrapolated norm ranges to
convert raw scores lying outside the range of standardization samples, whereas the new
calculations do make use of these. As a consequence, samples that would have
previously been given the minimum normed value (usually 60 or 65) are now given very
low values, even into the 40s. These extremely low values, of course, raise serious
questions about the measurement invariance of the tests (i.e. do they measure the same
thing in Germany as they do in Chad?), but it should be noted that some such values can
be expected on sampling error grounds alone. Since these are only found in developing
countries, especially African and Central American (e.g. Guatemala’s national IQ is
estimated to be 48), the use of the new more consistent methodology actually means
that Lynn had been overestimating
some non-European national IQs. In avoiding the use
of outside-norm values, Lynn effectively winsorised the low scoring samples, giving them
the benefit of the doubt. In dealing with these low scores, Lynn and Becker (2019)
recommend that authors apply a winsorization at 60 IQ in the book (p. 201). The
discussion of possible human bias in the prior calculations is surprisingly not given much
attention in the book at all, but Becker elaborated on it in a recent conference talk. He
analyzed relationships between changes in IQ estimates from the prior 2012 manual
calculations to the present mostly automated ones, and on relations to criterion variables
such as ancestry/ethnicity, and well-being. If such relationships are found it can be taken
to indicate a bias whereby the researcher pushes the calculation towards value that
better fit with other data. In Bayesian terms, this would actually be the right thing to do
(i.e. taking the prior into account), but it does open one to criticism if other researchers
have different priors (which of course they often do). Using the published dataset, I
calculated the correlations between European%, African%, HDI 2013, and the change in
scores between estimates. I estimated the ancestry fractions using Putterman’s
migration matrix. There were no relationships to ancestry: r European% = .11 (p =.17), r
African% = -.03 (p = .74). There was a weak relationship to HDI, r = .16 (p = .03),
meaning that lower HDI countries received slightly worse
scores in the new estimates,
opposite of the expected direction of bias. Thus, the criticism of bias against African
nations was contradicted. Furthermore, as can be seen in Figure 1, there is a marked
heteroscedasticity in the regression, i.e. the variance of changes to the scores is much
higher for the countries lower in intelligence and well-being. This of course reflects the
generally poorer data quality in these countries.
[fig. 1]
In reviews of the previous books in the series, much discussion centered on the use of
imputed country scores based on similar neighboring countries (Foster & Frijters, 2013;
Gale, 2013; Strate, 2013). In the new book, the same protocol is applied, but this time
using the border lengths as weights whenever possible (p. 43ff). Actually, data
imputation is not unusual in the least, and in fact should be the norm because the data
are not missing at random; indeed, missing values are concentrated in poorly developed
countries. If one used the national IQs without imputing the missing values, one would
get biased results from the missing observations (restriction of range bias, in particular).
Countries, of course, have spatial positions, and this allows the use of spatial statistics
(Gimond, 2019). Results from such analyses show that there is a very high degree of
(positive) spatial autocorrelation in the data, which in plain language means that country
IQs are highly predictable from neighboring countries (Gelade, 2008; Hassall & Sherratt,
2011). This, of course, also means that one can impute missing data with high accuracy,
justifying Lynn and Becker’s method. Spatial autocorrelation is just one type of
autocorrelation. Autocorrelation features are widely used to impute data in other fields,
for instance, in medicine where within person autocorrelation in time is used to fill in
missing data when observations in longitudinal studies are missing (Bell, Fiero, Horton, &
Hsu, 2014). Furthermore, if one used a multivariate imputation method, relationships to
other variables, such as health, ethnicity and wealth, would also be used to fill in missing
The relative stability of the national IQ estimates across decades of data compilation is
worth remarking on in more detail. Since about 2012, social science and biomedical
sciences have been plagued by open discussions about the lack of replication and
general unreliability of findings, called the replication crisis (Shrout & Rodgers, 2018). It
is generally agreed upon that one of the major causes of the poor reliability of findings is
that studies are too small and underpowered for their purposes. In contrast, as has been
remarked by Steven Pinker, there is little to no replication crisis in (non-candidate gene)
behavioral genetics, and IQ research generally (Pinker, 2015; Plomin, DeFries, Knopik, &
Neiderhiser, 2016). Reviews of statistical (median observed) power by field show why
this is the case. Nuijten, van Assen, Augusteijn, Crompvoets, & Wicherts (2018)
reviewed intelligence research for statistical power and found an overall power of 53%,
which compares favorably to other social science fields (e.g., neuroscience at 21%,
Button et al., 2013; economics, 18%, Ioannidis, Stanley, & Doucouliagos, 2017).
Furthermore, they looked at subfields of intelligence research and found that group
differences had a median power of 62%, the highest reported of any social science field,
and closing in on the minimal requirement of 80% suggested by Cohen decades ago.
The philosopher Neven Sesardic suggested an explanation for why this is so, namely
that because intelligence research and group difference research in particular is disliked
so much by generally left-wing academics (Duarte et al., 2015), the standards of
evidence in peer review have been increased (Sesardić, 2005, sec. 6.4). While this
results in some suppression of published works, it also has the effect of increasing the
average rigor of the published research. As a case in point, the median sample size of
studies in psychology is somewhere between 40 to 120 depending on subfield
(Marszalek, Barber, Kohlhart, & Holmes, 2011), whereas the median sample size in the
current national IQ dataset was 353.
The third chapter (116 pages) of the book relates to the correlations between national
IQs and various other variables. This chapter is structured like the previous books: each
section summarizes findings from studies using the national IQs. These sections do not
represent systematic reviews of studies published, but seem to be the authors’ chosen
examples. The values from the reviewed studies (i.e., the correlations) are also given in
tables. The presentation here is very dry. As other reviewers have noted, this was the
case also for the previous books: X study reported a correlation of A, Y study reported a
correlation of B and so on. There is little to no attempt at describing results from more
causally informative studies, such as time-lagged regressions, path models, or various
econometric designs. As such, it invites the skeptical reader to think that the authors are
simply assuming causation at the aggregate level from the correlations (Barnett &
Williams, 2004). The authors probably regard these as probable inferences based on the
strong evidence from individual level studies (Herrnstein & Murray, 1994; Strenze, 2015;
Trzaskowski et al., 2014), but they make no serious attempt at convincing a skeptical
reader, which is a pity. (For examples of studies which attempt to decompose cause and
effect, see e.g. (Christainsen, 2013; Jones, 2016; Jones & Potrafke, 2014; Jones &
Schneider, 2010; Rindermann, 2018; Wong, 2007).
A problem raised by the authors, but not adequately discussed, concerns the matter of
measurement bias. The omission is odd because the unestablished measurement
invariance of national IQs has been a frequent point of criticism (e.g., Wicherts &
Wilhelm, 2007). There are in fact a small number of studies that have examined
measurement invariance in cognitive ability more broadly on the national level. The first
question to be asked, perhaps, is whether a general factor of intelligence exists at all in
the data from poor non-Western countries. It’s conceivable that this factor could vary by
level of development and perhaps be smaller or absent in poor countries (though this
would be in contradiction of the so called Spearman’s law of diminishing returns (Blum &
Holling, 2017; see also Coyle & Rindermann, 2013)). The question was answered with a
large analysis by Warne & Burningham (2018). These authors found that almost every
dataset analyzed from non-Western countries showed a g-factor similar to that seen in
Western datasets. Further, one might ask whether methods for detecting measurement
invariance find that the tests function similarly across different countries. Research using
the international TIMSS mathematics test data by (Wu, Li, & Zumbo, 2007) analyzed
data from Western developed countries, as well as Northeast Asia (Japan, South Korea,
and Hong Kong). They used state of the art multi-group confirmatory factor analysis, and
found that scores were only comparable within the broad cultures, not across cultures.
That is, one could compare scores for e.g., USA and Australia, as well as Japan and
South Korea, but not e.g. for USA and Japan. Other research has also found that
measurement invariance held for comparisons between the USA and Canada (Bowden,
Saklofske, & Weiss, 2011). The lack of measurement invariance is concerning, and
means that one cannot simply interpret the score difference between nations as being of
the same nature as that between individuals with nations. The matter clearly calls for
further investigation, which can be done using the publicly available data in scholastic
ability datasets (PISA etc.), as well as the various translations of IQ batteries such as the
Wechsler batteries.
The fourth and final chapter discusses the future of national IQs. In fact, the chapter is
about how to increase national IQs, and takes for granted that these are to some extent
causal for the many associations summarized in the third chapter. The authors discuss
five ways: three environmental and two genetic. Nutrition is advocated as an important
cause, and the authors cite a few studies of breastfeeding and vitamin/mineral
supplementation. However, while one can find such studies, there are other equally or
better studies showing no effect. Moreover, the authors do not cite recent studies
examining breastfeeding’s effect on IQ using a sibling control design (which controls for
genetic confounding; Der, Batty, & Deary, 2006), nor a study with a very thorough set of
parental controls (Girard, Doyle, & Tremblay, 2017), both of which show negative effects.
A particularly good study is the randomized controlled trial of about 800 children in Nepal
whose mothers received or did not receive multi-vitamins while pregnant, and which
followed the children until age 12 when they were tested for IQ (Dulal et al., 2018). There
was only a 1 IQ advantage (p = .18) for the intervention group despite the well designed
intervention and the large sample size. Thus, changing nutrition to increase intelligence
in the way the authors propose is probably not as easy as their discussion implies. The
second environmental factor the authors suggest is improving health. While this may be
done on general well-being grounds, does it really improve intelligence? The authors cite
two studies of infectious diseases (Hadidjaja et al., 1998; Jardim-Botelho et al., 2008).
Unfortunately, neither are persuasive. The first is a cross-sectional study without rigorous
controls which is expected to have genetic confounding. The second is a randomized
controlled trial. However, the analysis and reporting is suboptimal, and it’s difficult to
work out what the effect size is; the post-test score means reported for traits of interest
are quite comparable between intervention and control groups. The authors had 6
measures of intelligence, but only 3 of them showed an effect (in ANOVA) despite a
comparatively large sample size of 483. While it is very likely that intelligence levels
could be increased by improving health, these particular studies do not provide strong
evidence for action. For a recent meta-analysis of interventions with somewhat optimistic
conclusions, see (Protzko, 2017). The third proposed method is improved education. Of
course, the relationship between education and intelligence is complicated, and still
unsettled. A recent meta-analysis of studies showed that the method employed to
estimate the causal effect of education on intelligence has a large effect on the estimated
outcome. Specifically, pre-post control studies finding quite small effects, while natural
experiments based on policy changes find, probably unrealistically, large effect sizes
(Ritchie & Tucker-Drob, 2018). The matter is also complicated by the fact that IQ gains
associated with education duration do not appear to be on the general intelligence factor
(g), but rather on the non-g factors (Ritchie, Bates, & Deary, 2015). Considering the
evidence that it’s mainly the g-factor that adds predictive validity to IQ test scores
(Jensen, 1998), it is not clear what improving non-g factor scores would accomplish. The
matter requires more psychometrically sophisticated research to clarify.
The authors also discuss genetic means of improving national intelligence. First, the
authors review of history of research into dysgenics, chiefly in studies reporting negative
relationships between IQ measures and fertility measures. Second, they review
economic policies attempting to encourage smart people to have more children,
especially smart women. They cite some old reviews of policies on maternity leave, and
tax/cash benefits in Western countries, finding small positive effects on fertility. Given
that Western countries vary widely in fertility levels, from near replacement (about 1.8) in
Nordic countries and the United Kingdom to quite low (about 1.4) in neighboring
Germany and southern Europe, it seems likely that one can indeed influence fertility
levels by interventions. Unfortunately, there seems to be a lack of randomized controlled
trials on the topic, so researchers are left with suboptimal research designs. Alternatively,
instead of attempting to get smart people to have more children, one could attempt to get
less bright people have fewer children. The most obvious way to do this, the authors
note, is to scale back welfare policies that enable the practice of single-motherhood. The
authors consider such changes to be doubtful in Western countries due to popular
resistance. Finally, the authors discuss the role of immigration in national IQs. Letting in
immigrants with lower IQs will generally lead to a decline in national IQ whereas letting in
high IQ immigrants will have the opposite effect (Borjas, 2016; Kirkegaard & Tranberg,
2015; Nyborg, 2012; Rindermann & Thompson, 2016; Woodley of Menie,
Peñaherrera-Aguirre, Fernandes, & Figueredo, 2018). They discuss the current political
realities of immigration, which need not concern us here. The final prediction for the
future echoes Lynn’s previous writings (Lynn, 2001): Western civilization is declining for
a variety of reasons, and China will probably emerge as the global superpower sometime
in this century.
All in all, the book is similar to the predecessors in presentation and discussion of
material. On the negative side, there is little to no attempt at using advanced statistical
methods to clarify matters of disputed causality, or even just the relative importance of
predictors. Existing studies on the question are not seriously discussed either, and an
important opportunity is missed. The presentation is quite dry. On the positive side, the
book describes the current state of the art calculations for national IQs, introduces the
reader to the open dataset of national IQs, shows their high replicability, and reviews
their use by tens if not hundreds of other researchers. This last point bears noting
because it shows that, despite criticism, the national IQs subfield has become a very
productive research program (Rindermann, 2013b; Urbach, 1974a, 1974b). In fact, one
might say that national IQs are getting quite popular because various other groups have
begun publishing very similar national cognitive ability estimates (Angrist, Djankov,
Goldberg, & Patrinos, 2019; Coutrot et al., 2018; Lim et al., 2018), even if they call it
something other than “intelligence”, and rarely cite the pioneering efforts of Richard Lynn
and colleagues.
References (Zotero)
Angrist, N., Djankov, S., Goldberg, P., & Patrinos, H. A. (2019). Measuring Human
(SSRN Scholarly Paper No. ID 3339416). Retrieved from Social Science
Research Network website:
Barnett, S. M., & Williams, W. (2004). National Intelligence and The Emperor’s New
. 49
(4), 389–396.
Bell, M. L., Fiero, M., Horton, N. J., & Hsu, C.-H. (2014). Handling missing data in RCTs;
a review of the top medical journals. BMC Medical Research Methodology
, 14
Blum, D., & Holling, H. (2017). Spearman’s law of diminishing returns. A meta-analysis.
, 65
, 60–66.
Borjas, G. J. (2016). We wanted workers: unraveling the immigration narrative
edition). New York: W. W. Norton & Company.
Bowden, S. C., Saklofske, D. H., & Weiss, L. G. (2011). Augmenting the Core Battery
With Supplementary Subtests: Wechsler Adult Intelligence Scale—IV
Measurement Invariance Across the United States and Canada. Assessment
(2), 133–140.
Button, K. S., Ioannidis, J. P. A., Mokrysz, C., Nosek, B. A., Flint, J., Robinson, E. S. J.,
& Munafò, M. R. (2013). Power failure: why small sample size undermines the
reliability of neuroscience. Nature Reviews Neuroscience
, 14
(5), 365–376.
Christainsen, G. B. (2013). IQ and the wealth of nations: How much reverse causality?
, 41
(5), 688–698.
Coutrot, A., Silva, R., Manley, E., Cothi, W. de, Sami, S., Bohbot, V. D., … Spiers, H. J.
(2018). Global Determinants of Navigation Ability. Current Biology
, 28
Coyle, T. R., & Rindermann, H. (2013). Spearman’s Law of Diminishing Returns and
national ability. Personality and Individual Differences
, 55
(4), 406–410.
Der, G., Batty, G. D., & Deary, I. J. (2006). Effect of breast feeding on intelligence in
children: prospective study, sibling pairs analysis, and meta-analysis. BMJ
(7575), 945.
Duarte, J. L., Crawford, J. T., Stern, C., Haidt, J., Jussim, L., & Tetlock, P. E. (2015).
Political diversity will improve social psychological science. Behavioral and Brain
, 38
Dulal, S., Liégeois, F., Osrin, D., Kuczynski, A., Manandhar, D. S., Shrestha, B. P., …
Prost, A. (2018). Does antenatal micronutrient supplementation improve
children’s cognitive function? Evidence from the follow-up of a double-blind
randomised controlled trial in Nepal. BMJ Global Health
, 3
(1), e000527.
Flynn, J. R. (1984). The mean IQ of Americans: Massive gains 1932 to 1978.
Psychological Bulletin
, 95
(1), 29–51.
Foster, G., & Frijters, P. (2013). Intelligence: A Unifying Construct for the Social
Sciences, Richard Lynn, Tatu Vanhanen. Ulster Institute for Social Research
(2012). xiv + 530 pp., ISBN: 978-0-9568811-8-2. Journal of Economic
, 39
, 439–440.
Gale, C. R. (2013). Review of: Intelligence. A unifying construct for the social sciences,
Lynn, R., & Vanhanen, T., London: Ulster Institute for Social Research, ISBN
978-0-9568811-8-2, 530 pp., (introductory material xiv). Intelligence
, 41
Gelade, G. A. (2008). The geography of IQ. Intelligence
, 36
(6), 495–501.
Gimond, M. (2019). Intro to GIS and Spatial Analysis
. Retrieved from
Girard, L.-C., Doyle, O., & Tremblay, R. E. (2017). Breastfeeding, Cognitive and
Noncognitive Development in Early Childhood: A Population Study. Pediatrics
Hadidjaja, P., Bonang, E., Suyardi, M. A., Abidin, S. A., Ismid, I. S., & Margono, S. S.
(1998). The effect of intervention methods on nutritional status and cognitive
function of primary school children infected with Ascaris lumbricoides. The
American Journal of Tropical Medicine and Hygiene
, 59
(5), 791–795.
Hassall, C., & Sherratt, T. N. (2011). Statistical inference and spatial patterns in
correlates of IQ. Intelligence
, 39
(5), 303–310.
Herrnstein, R. J., & Murray, C. (1994). The Bell Curve: Intelligence and Class Structure
in American Life
. New York: Free Press.
Hesse, B. W. (2018). Can psychology walk the walk of open science? American
, 73
(2), 126–137.
Ioannidis, J. P. A., Stanley, T. D., & Doucouliagos, H. (2017). The Power of Bias in
Economics Research. The Economic Journal
, 127
(605), F236–F265.
Jardim-Botelho, A., Raff, S., Rodrigues, R. de A., Hoffman, H. J., Diemert, D. J.,
Corrêa-Oliveira, R., … Gazzinelli, M. F. (2008). Hookworm, Ascaris lumbricoides
infection and polyparasitism associated with poor cognitive performance in
Brazilian schoolchildren. Tropical Medicine & International Health: TM & IH
(8), 994–1004.
Jensen, A. R. (1998). The g factor: the science of mental ability
. Westport, Conn.:
Jones, G. (2016). Hive mind: how your nation’s IQ matters so much more than your own
Stanford, California: Stanford Economics and Finance, an imprint of Stanford
University Press.
Jones, G., & Potrafke, N. (2014). Human capital and national institutional quality: Are
TIMSS, PISA, and national average IQ robust predictors? Intelligence
, 46
Jones, G., & Schneider, W. J. (2010). IQ in the Production Function: Evidence from
Immigrant Earnings. Economic Inquiry
, 48
(3), 743–755.
Kirkegaard, E. O. W., & Tranberg, B. (2015). Increasing inequality in general intelligence
and socioeconomic status as a result of immigration in Denmark 1980-2014 |
Retrieved from
Lim, S. S., Updike, R. L., Kaldjian, A. S., Barber, R. M., Cowling, K., York, H., … Murray,
C. J. L. (2018). Measuring human capital: a systematic analysis of 195 countries
and territories, 1990–2016. The Lancet
, 392
(10154), 1217–1234.
Lindsay, D. S. (2017). Sharing Data and Materials in Psychological Science.
Psychological Science
, 28
(6), 699–702.
Lynn, R. (1978). Ethnic and racial differences in intelligence: International comparisons.
In Human Variation: The Biopsychology of Age, Race, and Sex
. Academic Press:
New York, NY, USA.
Lynn, R. (2001). Eugenics: a reassessment
. Westport, Conn: Praeger.
Lynn, R., & Vanhanen, T. (2002). IQ and the wealth of nations
. Westport, Conn: Praeger.
Lynn, R., & Vanhanen, T. (2006). IQ and global inequality
. Augusta, Ga: Washington
Summit Publishers.
Lynn, R., & Vanhanen, T. (2012). Intelligence: A Unifying Construct for the Social
(1st ed.). Ulster Institute for Social Research.
Marszalek, J. M., Barber, C., Kohlhart, J., & Holmes, C. B. (2011). Sample size in
psychological research over the past 30 years. Perceptual and Motor Skills
(2), 331–348.
Nuijten, M. B., van Assen, M. A. L. M., Augusteijn, H., Crompvoets, E. A. V., & Wicherts,
J. M. (2018). Effect Sizes, Power, and Biases in Intelligence Research: A
Nyborg, H. (2012). The decay of Western civilization: Double relaxed Darwinian
Selection. Personality and Individual Differences
, 53
(2), 118–125.
Paxton, A., & Tullett, A. (2019). Open Science in Data-Intensive Psychology and
Cognitive Science. Policy Insights from the Behavioral and Brain Sciences
, 6
Pinker, S. (2015, September 19). Irony: Replicability crisis in psych DOESN’T apply to
IQ: huge n’s, replicable results. But people hate the message
. Retrieved from
Plomin, R., DeFries, J. C., Knopik, V. S., & Neiderhiser, J. M. (2016). Top 10 Replicated
Findings From Behavioral Genetics. Perspectives on Psychological Science: A
Journal of the Association for Psychological Science
, 11
(1), 3–23.
Protzko, J. (2017). Raising IQ among school-aged children: Five meta-analyses and a
review of randomized controlled trials. Developmental Review
, 46
, 81–101.
Rindermann, H. (2007). The g-factor of international cognitive ability comparisons: the
homogeneity of results in PISA, TIMSS, PIRLS and IQ-tests across nations.
European Journal of Personality
, 21
(5), 667–706.
Rindermann, H. (2013a). African cognitive ability: Research, results, divergences and
recommendations. Personality and Individual Differences
, 55
(3), 229–233.
Rindermann, H. (2013b). The Intelligence of Nations: A Productive Research
Paradigm—Comment on Hunt (2012). Perspectives on Psychological Science
(2), 190–192.
Rindermann, H. (2018). Cognitive capitalism: human capital and the wellbeing of nations
Cambridge, United Kingdom; New York, NY: University Printing House.
Rindermann, H., & Thompson, J. (2016). THE COGNITIVE COMPETENCES OF
OF GAPS, POSSIBLE CAUSES AND IMPACT. Journal of Biosocial Science
(1), 66–93.
Ritchie, S. J., Bates, T. C., & Deary, I. J. (2015). Is education associated with
improvements in general cognitive ability, or in specific skills? Developmental
, 51
(5), 573–582.
Ritchie, S. J., & Tucker-Drob, E. M. (2018). How Much Does Education Improve
Intelligence? A Meta-Analysis. Psychological Science
, 29
(8), 1358–1369.
Sesardić, N. (2005). Making sense of heritability
. Retrieved from
Shrout, P. E., & Rodgers, J. L. (2018). Psychology, Science, and Knowledge
Construction: Broadening Perspectives from the Replication Crisis. Annual
Review of Psychology
, 69
(1), 487–510.
Smith, A. (1776). An inquiry into the nature and causes of the wealth of nations
. Chicago:
University of Chicago Press.
Strate, J. M. (2013). Richard Lynn and Tatu Vanhanen, Intelligence: A Unifying Construct
for the Social Sciences (London: Ulster Institute for Social Research, 2012), 552
pages. ISBN 978-0-9568811-8-2. Hardcover $65.00. Politics and the Life
, 32
(1), 108–110.
Strenze, T. (2015). Intelligence and Success. In S. Goldstein, D. Princiotta, & J. A.
Naglieri (Eds.), Handbook of Intelligence
(pp. 405–413). Retrieved from
Swartz, M. K. (2011). The PRISMA statement: a guideline for systematic reviews and
meta-analyses. Journal of Pediatric Health Care: Official Publication of National
Association of Pediatric Nurse Associates & Practitioners
, 25
(1), 1–2.
Trzaskowski, M., Harlaar, N., Arden, R., Krapohl, E., Rimfeld, K., McMillan, A., …
Plomin, R. (2014). Genetic influence on family socioeconomic status and
children’s intelligence. Intelligence
, 42
, 83–88.
Urbach, P. (1974a). Progress and Degeneration in the “IQ Debate” (I). The British
Journal for the Philosophy of Science
, 25
(2), 99–135. Retrieved from JSTOR.
Urbach, P. (1974b). Progress and Degeneration in the “IQ Debate” (II). The British
Journal for the Philosophy of Science
, 25
(3), 235–259. Retrieved from JSTOR.
Warne, R. T., & Burningham, C. (2018). Spearman’s g Found in 31 Non-Western
Cultures: Strong Evidence that g is a Universal Trait. PsyArXiv
Wicherts, J. M., Dolan, C. V., & van der Maas, H. L. J. (2010). A systematic literature
review of the average IQ of sub-Saharan Africans. Intelligence
, 38
(1), 1–20.
Wicherts, J. M., & Wilhelm, O. (2007). What is the national g-factor? European Journal of
, 21
(5), 763–765.
Wong, B. (2007). Cognitive Ability (IQ), Education Quality, Economic Growth, Human
Migration: Implications from a Sociobiological Paradigm of Global Economic
Inequality. Mankind Quarterly
, 48
(1). Retrieved from
Woodley of Menie, M. A., Peñaherrera-Aguirre, M., Fernandes, H. B. F., & Figueredo,
A.-J. (2018). What causes the anti-Flynn effect? A data synthesis and analysis of
predictors. Evolutionary Behavioral Sciences
, 12
(4), 276–295.
Wu, A. D., Li, Z., & Zumbo, B. D. (2007). Decoding the Meaning of Factorial Invariance
and Updating the Practice of Multi-group Confirmatory Factor Analysis&58; A
Demonstration With TIMSS Data [Text]. Retrieved April 17, 2019, from
... Many other sets of estimates have been produced by others, most importantly Heiner Rindermann (2018). However, we used the 2012 dataset because, as of writing, it is more comprehensive than the 2019 recalculation (see discussion in Kirkegaard, 2019c). For estimates of the proportion of Muslims in each group, we used estimates from Pew Research (Pew Research Center, 2011). ...
... His work is described in a recent book coauthored with Lynn (Lynn & Becker, 2019). Generally speaking, the estimates can be considered quite reliable for many countries, but not all, and much work remains to be done examining questionable aspects of measurement invariance (Dutton et al., 2018;Kirkegaard, 2019c). ...
Full-text available
We estimated crime rates among 70 origin-based immigrant groups in the Netherlands for the years 2005-2018. Results indicated that crime rates have overall been falling for each group in the period of study, and in the country as a whole, with about a 50% decline since 2005. Immigrant groups varied widely in crime rates, with East Asian countries being lower and Muslim countries, as well as Dutch (ex-)colonial possessions in the Caribbean, being higher than the natives. We found that national IQ and Muslim percentage of population of the origin countries predicted relative differences in crime rates, r’s = .64 and .45, respectively, in line with previous research both in the Netherlands and in other European countries. Furthermore, we carried out a survey of 200 persons living in the Netherlands to measure their preferences for immigrants for each origin country in terms of getting more or fewer persons from each country. Following Carl (2016), we computed a mean opposition metric for each country. This correlated strongly with actual crime rates we found, r’s = .46 and .57, for population weighted and unweighted results, respectively. The main outliers in the regression were the Dutch (ex-)colonial possessions, and if these are excluded, the correlations increase to .68 and .66, respectively. Regressions with plausible confounders (Muslim percentage, geographical fixed effects) showed that crime rates continued to be a useful predictor of opposition to specific countries. The results were interpreted as being in line with a rational voter preference for less crime-prone immigrants.
... While this should be seen in the light of the so-called American woke cultural revolution (E. Kaufman, 2020), it is nevertheless true that it is difficult to accurately estimate average national intelligence owing to data limitations (Kirkegaard, 2019b;Lynn & Becker, 2019;Rindermann, 2013). For many countries of the world, chiefly in poor regions, there simply are no good data on measured cognitive ability. ...
Full-text available
Immigrants to Western countries typically have worse social outcomes than natives, but country of origin immigrant groups differ widely. We studied school performance in Denmark for 116 immigrant groups measured by the grade point average (GPA) of the 9th grade exam at the end of compulsory schooling. General intelligence is a strong causal factor of school outcomes and life outcomes in general for individuals. We accordingly predicted that country of origin average intelligence (national IQ) will predict immigrant group outcomes. We furthermore included as covariates immigrant generation (first vs. second) as well as the Muslim percentage of country of origin. Results show that GPA in Denmark can be predicted by national IQ r = .47 (n = 81), Muslim percentage r = -.40 (n = 81), and educational selectivity of immigrants entering Denmark r = .35 (n = 71). Regression modeling indicated that each predictor is informative when combined. The final model explained 46.3% of the variance with first generation (binary) β = -0.65, βIQ = 0.29, βMuslim = -0.21, and β education selectivity index = 0.27 (all predictors p < .001, n = 97). Our results are in line with existing research on cognitive stratification and immigration.
... (current version v1.3.3), and has also been made available in book form which provides an updated review of the various published findings related to national IQ (Lynn & Becker, 2019). However, even though the Becker dataset is more methodologically rigorous, it covers fewer studies thus far (about 70% of the Lynn 2012 database), and thus has less reliable and fewer estimates (Kirkegaard, 2019b). Thus, in some ways, the 2012 Lynn dataset is still superior for empirical analysis. ...
Full-text available
Patient people fare better in life than impatient people. Based on this and on economic models, many economists have claimed that more patient countries should fare better than less patient countries. We utilize cross-national data in non-cognitive traits measured in the Global Preference Survey (GPS). This survey measured six non-cognitive traits — risk and time preferences, positive and negative reciprocity, altruism, and trust — across 76 countries in about 80,000 persons. As such, it provides the best current database of economics-focused non-cognitive traits. We combine this database with existing estimates of national intelligence (national IQs) and model country outcomes as a function of these predictors. For outcomes, we used the 51 national well-being indicators from the Social Progress Index (SPI) as well as the composite extracted from this, the general socioeconomic factor. We find that non-cognitive variables, time preference included, are only weakly predictive of national well-being outcomes when national IQs are also in the model. The median β across the indicators was 0.11 for time preference but 0.39 for national IQ. We replicated these results using six economic indicators, again with similar results: median βs of 0.15 and 0.52 for time preference and national IQ, respectively. Across all our results, we found that national IQ has 2-4 times the predictive validity of time preference. These results are fairly robust to inclusion of a spatial autocorrelation control, alternative measures of national IQ and time preference, or no controls. Our results suggest that the importance of national non-cognitive traits, including time preference, is overestimated or that these traits are mismeasured.
Intelligence research has been subjected to rigorous scientific examination, but it also faces open hostility from opponents within psychology, from academic disciplines outside of psychology, and from the media. This does not occur for highly technical research topics (such as psychometrics), but seems to apply particularly to group differences research. Apart from often assumed harm to scientific research, political and societal pressures may paradoxically also lead to epistemically stronger theories in the long run. Extending insights from Neven Sesardic (Making sense of heritability. Cambridge University Press, 2005), it is hypothesized that such pressures lead not only to stronger theories but to a stronger research field and to positive selection effects regarding the depth and breadth of intellectual interests and abilities of researchers. There also may be selection effects for personality traits (e.g., more ego strength, originality, and bizarreness), as well as more cooperativeness and solidarity among intelligence researchers. Several quantitative indicators demonstrate such effects (e.g., higher power of empirical studies within intelligence research compared to other fields and higher average publication counts and h-indexes). In case studies, the behavior of several intelligence researchers is described, which substantiates assumed personality effects.
Full-text available
Previous studies have found that the ability of a country’s cognitive elite is generally more predictive than the average ability. However, these studies have relied on sub-optimal methods. Here, the authors tested smart fraction theory, as it is known, using a pre-residualization approach, which obviates the problem of collinearity. For outcome variables, they utilised the 51 indicators of the Social Progress Index, as well as 6 economic variables. Like in previous research, the authors operationalized the ability of the intellectual class as the 95th percentile score. Consistent with smart fraction theory, they found evidence that the ability of the intellectual class does influence outcomes over and above that of the average ability. For the 93 countries with available data, average ability explained 57% of the variation in country performance. Adding the 95th percentile score increased this to 66%. The authors discuss their findings in light of the existing literature.
Full-text available
Background: Human capital is recognised as the level of education and health in a population and is considered an important determinant of economic growth. The World Bank has called for measurement and annual reporting of human capital to track and motivate investments in health and education and enhance productivity. We aim to provide a new comprehensive measure of human capital across countries globally. Methods: We generated a period measure of expected human capital, defined for each birth cohort as the expected years lived from age 20 to 64 years and adjusted for educational attainment, learning or education quality, and functional health status using rates specific to each time period, age, and sex for 195 countries from 1990 to 2016. We estimated educational attainment using 2522 censuses and household surveys; we based learning estimates on 1894 tests among school-aged children; and we based functional health status on the prevalence of seven health conditions, which were taken from the Global Burden of Diseases, Injuries, and Risk Factors Study 2016 (GBD 2016). Mortality rates specific to location, age, and sex were also taken from GBD 2016. Findings: In 2016, Finland had the highest level of expected human capital of 28·4 health, education, and learning-adjusted expected years lived between age 20 and 64 years (95% uncertainty interval 27·5-29·2); Niger had the lowest expected human capital of less than 1·6 years (0·98-2·6). In 2016, 44 countries had already achieved more than 20 years of expected human capital; 68 countries had expected human capital of less than 10 years. Of 195 countries, the ten most populous countries in 2016 for expected human capital were ranked: China at 44, India at 158, USA at 27, Indonesia at 131, Brazil at 71, Pakistan at 164, Nigeria at 171, Bangladesh at 161, Russia at 49, and Mexico at 104. Assessment of change in expected human capital from 1990 to 2016 shows marked variation from less than 2 years of progress in 18 countries to more than 5 years of progress in 35 countries. Larger improvements in expected human capital appear to be associated with faster economic growth. The top quartile of countries in terms of absolute change in human capital from 1990 to 2016 had a median annualised growth in gross domestic product of 2·60% (IQR 1·85-3·69) compared with 1·45% (0·18-2·19) for countries in the bottom quartile. Interpretation: Countries vary widely in the rate of human capital formation. Monitoring the production of human capital can facilitate a mechanism to hold governments and donors accountable for investments in health and education. Funding: Institute for Health Metrics and Evaluation.
Full-text available
This study was conducted to investigate body mass index (BMI), levels of cholesterol and triglycerides in prison inmates at the Institution for Reform and Rehabilitation in Southern Libya to be considered as an indication about their health and the provided foods. The results of this study showed that 26.5% of BMI of the prison inmates were found to be higher than the normal levels. Generally, the average level of cholesterol and triglycerides concentrations were found to be within normal range 142.6 mg/dl and 135.4 mg/dl, respectively. The findings also established that there were a significant relationship and direct correlation between BMI levels and age and concentration of cholesterol and triglycerides levels. The results of this showed that the served foods for these prison inmates are well balanced as indicated by their cholesterol and triglycerides levels.
Full-text available
Human spatial ability is modulated by a number of factors, including age [1-3] and gender [4, 5]. Although a few studies showed that culture influences cognitive strategies [6-13], the interaction between these factors has never been globally assessed as this requires testing millions of people of all ages across many different countries in the world. Since countries vary in their geographical and cultural properties, we predicted that these variations give rise to an organized spatial distribution of cognition at a planetary-wide scale. To test this hypothesis, we developed a mobile-app-based cognitive task, measuring non-verbal spatial navigation ability in more than 2.5 million people and sampling populations in every nation state. We focused on spatial navigation due to its universal requirement across cultures. Using a clustering approach, we find that navigation ability is clustered into five distinct, yet geographically related, groups of countries. Specifically, the economic wealth of a nation was predictive of the average navigation ability of its inhabitants, and gender inequality was predictive of the size of performance difference between males and females. Thus, cognitive abilities, at least for spatial navigation, are clustered according to economic wealth and gender inequalities globally, which has significant implications for cross-cultural studies and multi-center clinical trials using cognitive testing.
Full-text available
Introduction Multiple Micronutrient (MMN) supplementation during pregnancy can decrease the proportion of infants born low birth weight and small for gestational age. Supplementation could also enhance children’s cognitive function by improving access to key nutrients during fetal brain development and increasing birth weight, especially in areas where undernutrition is common. We tested the hypothesis that children whose mothers received MMN supplementation during pregnancy would have higher intelligence in early adolescence compared with those receiving Iron and Folic Acid (IFA) only. Methods We followed up children in Nepal, whose mothers took part in a double-blind Randomised Controlled Trial (RCT) that compared the effects on birth weight and gestational duration of antenatal MMN versus IFA supplementation. We assessed children’s Full Scale Intelligence Quotient (FSIQ) using the Universal Non-verbal Intelligence Test (UNIT), and their executive function using the counting Stroop test. The parent trial was registered as ISRCTN88625934. Results We identified 813 (76%) of the 1069 children whose mothers took part in the parent trial. We found no differences in FSIQ at 12 years between MMN and IFA groups (absolute difference in means (diff): 1.25, 95% CI −0.57 to 3.06). Similarly, there were no differences in mean UNIT memory (diff: 1.41, 95% CI −0.48 to 3.30), reasoning (diff: 1.17, 95% CI −0.72 to 3.06), symbolic (diff: 0.97, 95% CI −0.67 to 2.60) or non-symbolic quotients (diff: 1.39, 95% CI −0.60 to 3.38). Conclusion Our follow-up of a double-blind RCT in Nepal found no evidence of benefit from antenatal MMN compared with IFA for children’s overall intelligence and executive function at 12 years.
Full-text available
An “open science movement” is gaining traction across many disciplines within the research enterprise but is also precipitating consternation among those who worry that too much disruption may be hampering professional productivity. Despite this disruption, proponents of open data collaboration have argued that some of the biggest problems of the 21st century need to be solved with the help of many people and that data sharing will be the necessary engine to make that happen. In the United States, a national strategic plan for data sharing encouraged the federally funded scientific agencies to (a) publish open data for community use in discoverable, machine-readable, and useful ways; (b) work with public and civil society organizations to set priorities for data to be shared; (c) support innovation and feedback on open data solutions; and (d) continue efforts to release and enhance high-priority data sets funded by taxpayer dollars. One of the more visible open data projects in the psychological sciences is the presidentially announced “Brain Research Through Advancing Innovative Neurotechnologies” (BRAIN) initiative. Lessons learned from initiatives such as these are instructive both from the perspective of open science within psychology and from the perspective of understanding the psychology of open science. Recommendations for creating better pathways to “walk the walk” in open science include (a) nurturing innovation and agile learning, (b) thinking outside the paradigm, (c) creating simplicity from complexity, and (d) participating in continuous learning evidence platforms.
Today, researchers can collect, analyze, and share more data than ever before. Not only does increasing technological capacity open the door to new data-intensive perspectives in cognitive science and psychology (i.e., research that takes advantage of complex or large-scale data to understand human cognition and behavior), but increasing connectedness has sparked exponential increases in the ease and practice of scientific transparency. The growing open science movement encourages researchers to share data, materials, methods, and publications with other scientists and the wider public. Open science benefits data-intensive psychological science, the public, and public policy, and we present recommendations to improve the adoption of open science practices by changing the academic incentive structure and by improving the education pipeline. Despite ongoing questions about implementing open science guidelines, policy makers have an unprecedented opportunity to shape the next frontier of scientific discovery.
We analyzed 2,439 effect sizes from 131 meta-analyses in intelligence research to estimate the average effect size, median power, and evidence for bias in this field. We found that the typical effect size in this field was a Pearson’s correlation of .26, and the median sample size was 60. We calculated the power of each primary study by using the corresponding meta-analytic effect as a proxy for the true effect. The median power across all studies was 48.8%, with only 29.8% of the studies reaching a power of 80% or higher. We documented differences in average effect size and median power between different subfields in intelligence research (correlational studies, studies of group differences, experiments, toxicology, and behavior genetics). Across all meta-analyses, we found evidence for small study effects in meta-analyses, highlighting potential publication bias. The evidence for the small study effect being stronger for studies from the US than for non-US studies (a US effect) was weak at best. We found no clear evidence for the decline effect, early extremes effect, or citation bias across meta-analyses. Even though the power in intelligence research seems to be higher than in other fields of psychology, this field does not seem immune to the problems of replicability as documented in psychology.
Intelligence test scores and educational duration are positively correlated. This correlation could be interpreted in two ways: Students with greater propensity for intelligence go on to complete more education, or a longer education increases intelligence. We meta-analyzed three categories of quasiexperimental studies of educational effects on intelligence: those estimating education-intelligence associations after controlling for earlier intelligence, those using compulsory schooling policy changes as instrumental variables, and those using regression-discontinuity designs on school-entry age cutoffs. Across 142 effect sizes from 42 data sets involving over 600,000 participants, we found consistent evidence for beneficial effects of education on cognitive abilities of approximately 1 to 5 IQ points for an additional year of education. Moderator analyses indicated that the effects persisted across the life span and were present on all broad categories of cognitive ability studied. Education appears to be the most consistent, robust, and durable method yet to be identified for raising intelligence.
Cambridge Core - Cognition - Cognitive Capitalism - by Heiner Rindermann