Academic journal article Mankind Quarterly

Non-Verbal Ability of Zulu Children across 56 Years

Academic journal article Mankind Quarterly

Non-Verbal Ability of Zulu Children across 56 Years

Article excerpt

A research program to collect IQs for all peoples and nations in the world was initiated by Lynn (1978) and has been updated in subsequent publications. In this research program population IQs are expressed in relation to a British mean of 100 and standard deviation of 15. They are described as "British IQs" or "Greenwich IQs", a term proposed by Rindermann (2012), analogous to the measurement of longitude which is set as a deviation from zero through Greenwich. Because performance on cognitive tests has increased over time in Britain as in other western countries, IQs are calculated according to British norms at the time a study was performed.

In this paper we contribute to this research program by reporting data on the non-verbal intelligence of Zulu primary school children in the South African province of KwaZulu-Natal. The province has a population of approximately 8 million, of which 60% is rural. Black Africans comprise 82% of the population today, with the largest proportion being Zulus. Pietermaritzburg is the capital city of the province, and Durban is the largest city.

In their most recent compilation, Lynn and Vanhanen (2012, p. 412) summarize 25 studies of the intelligence of black South Africans, with a range of British IQs between 65 and 82 and a median of 71. In the present study, we compare the performance of Zulu children from KwaZulu-Natal on tests of non-verbal abilities. Results from four studies performed up to 56 years apart are analyzed.

Study 1

Notcutt (1949) reported the results of a study performed in 1948 in which Raven's Standard Progressive Matrices (SPM) test was administered to 762 Zulu primary school pupils aged 11 to 15 years in the vicinity of Durban. The children are described as "living under urban and peri-urban conditions." The SPM is a non-verbal test of inductive reasoning in which subjects have to identify the missing piece in a sequence of visual patterns. The easiest items require primarily visual pattern matching while the more difficult ones require the detection of regularities based on abstract properties such as symmetry, numerosity and directionalty. Therefore the test is thought to discriminate mainly on visual-perceptual ability in low-scoring samples and on non-verbal reasoning ability in high-scoring samples.

Although the Raven tests are described as "culture-reduced," there have been large performance increases ("Flynn effects") on these tests in Western countries during the 20th century (Flynn, 2007, 2009; Lynn, 2009; Raven, 2008; Raven, Raven & Court, 1998). These gains indicate that performance on these tests does respond to changing environmental conditions across time.

The results of the Notcutt (1949) study are summarized in Table 1. The table also shows the IQ equivalents of the scores according to British norms derived from standardizations in 1943 (Raven, 1960) and 1979 (Raven, Raven and Court, 1998). The higher scores on the 1943 than the 1979 norms indicate the IQ gain that was taking place in Britain during this interval. The last column gives the British IQ in 1948 as the time-weighted average calculated from the 1943 and 1979 IQs.

Study 2

In another early study,Cowleyand Murray (1962) reported data on Piagetian tests for 40 white and 40 Zulu children matched for age and aged 5-12 years, with equal numbers of boys and girls. The children were in schools in townships in KwaZulu-Natal. Three Piagetian tests were administered including tests of topological space (drawings of shapes), projective space (drawings of perspective), and Euclidian space (identification of shapes). The authors write that "The children's reactions were classified into stages of development ...based on that of Piaget." Means and standard deviations are given for the three tests for the two groups. The white children scored significantly higher on all three tests (p<.001). The differences expressed in standard deviation units were 1.30, 1. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.