Editing Race and intelligence (section)

=== Spearman's hypothesis ===
{{Main|Spearman's hypothesis}}
Spearman's hypothesis states that the magnitude of the black–white difference in tests of cognitive ability depends entirely or mainly on the extent to which a test measures general mental ability, or ''g''. The hypothesis was first formalized by [[Arthur Jensen]], who devised the statistical "method of correlated vectors" to test it. If Spearman's hypothesis holds true, then the cognitive tasks that have the highest ''g''-load are the tasks in which the gap between black and white test takers are greatest. Jensen and Rushton took this to show that the cause of ''g'' and the cause of the gap are the same—in their view, genetic differences.{{sfn|Rushton|Jensen|2005}}

{{harvtxt|Mackintosh|2011|pages=338–39}} acknowledges that Jensen and Rushton showed a modest correlation between ''g''-loading, heritability, and the test score gap, but does not agree that this demonstrates a genetic origin of the gap. Mackintosh argues that it is exactly those tests that Rushton and Jensen consider to have the highest ''g''-loading and heritability, such as the Wechsler test, that have seen the greatest increases in black performance due to the Flynn effect. This likely suggests that they are also the most sensitive to environmental changes, which undermines Jensen's argument that the black–white gap is most likely caused by genetic factors. {{harvtxt|Nisbett|Aronson|Blair|Dickens|2012a|page=146}} make the same point, noting also that the increase in the IQ scores of black test takers necessarily indicates an increase in ''g''.

James Flynn argued that his findings undermine Spearman's hypothesis.<ref>{{Cite journal |last1=Flynn |first1=J.R. |year=1999 |title=Searching for justice: the discovery of IQ gains over time |url=http://www.stat.columbia.edu/~gelman/stuff_for_blog/flynn.pdf |url-status=live |journal=American Psychologist |volume=54 |pages=5–9 |doi=10.1037/0003-066X.54.1.5 |archive-url=https://web.archive.org/web/20100625085640/http://www.stat.columbia.edu/~gelman/stuff_for_blog/flynn.pdf |archive-date=25 June 2010 |access-date=26 October 2017}}</ref> In a 2006 study, he and William Dickens found that between 1972 and 2002 "The standard measure of the ''g'' gap between Blacks and Whites declined virtually in tandem with the IQ gap."{{sfn|Dickens|Flynn|2006}} Flynn also criticized Jensen's basic assumption that a correlation between ''g''-loading and test score gap implies a genetic cause for the gap.<ref>{{cite journal |last=Flynn |first=James R. |year=2010 |title=The spectacles through which I see the race and IQ debate |url=http://www.iapsych.com/iqmr/fe/LinkedDocuments/flynn2010a.pdf |journal=Intelligence |volume=38 |issue=4 |pages=363–366 |doi=10.1016/j.intell.2010.05.001 |access-date=2011-02-18 |archive-date=2020-12-07 |archive-url=https://web.archive.org/web/20201207224050/http://www.iapsych.com/iqmr/fe/LinkedDocuments/flynn2010a.pdf |url-status=live }}</ref> In a 2014 suite of meta-analyses, along with co-authors Jan te Nijenhuis and Daniel Metzen, he showed that the same negative correlation between IQ gains and ''g''-loading obtains for cognitive deficits of known environmental cause: [[iodine deficiency]], [[prenatal cocaine exposure]], [[Fetal alcohol spectrum disorder|fetal alcohol syndrome]], and [[traumatic brain injury]].<ref>{{Cite journal |last1=Flynn |first1=James R. |last2=te Nijenhuis |first2=Jan |last3=Metzen |first3=Daniel |date=2014 |title=The g beyond Spearman's g: Flynn's paradoxes resolved using four exploratory meta-analyses |url=https://james-flynn.net/wp-content/uploads/2020/12/flynn2014-The-g-beyond-Spearmans-g-Flynns-paradoxes-resolved-using-four-exploratory-meta-analyses.pdf |journal=Intelligence |volume=44 |pages=1–10|doi=10.1016/j.intell.2014.01.009 }}</ref>