De intelligence te Amsterdam in verband met demografische en sociologische kenmerken van de bevolking

Intelligence in Amsterdam in connection with demographic and sociological characteristics of the population. Since the end of the war Dutch recruits, after having passed a medical examination, have been submitted to five psychological tests, measuring general intelligence, aptitude for mathematics and languages, respectively, technical ability and ability to carry out orders. This investigation is based on the results of some 13,700 recruits examined in Amsterdam from 1947 up to and including 1949. It is restricted to the results of the test for general intelligence, as this test is supposed to be most free from environmental influences and as its results are available for all recruits. General intelligence has been measured by the test of progressive matrices devised by f. C. Raven. The correlation matrix for the five tests (based on approximately 9,500 recruits) has been calculated and although the correlation coefficients are all positive and highly significant, they vary between 52 and 76; showing that the possibility cannot be excluded of finding greatly deviating results when using the results of one of the other tests. In order to be able to compare intelligence with characteristics not available for the individual recruits, either as a consequence of lacking statistical data, or even due to logical impossibility (e.g. in the case of fertility), the average level of intelligence has been calculated for each of the 54 wards of Amsterdam for which the Amsterdam Municipal Bureau of Statistics regularly compiles statistical data. Consequently the assumption has been made that these averages may be considered as fair measures for the average intelligence of the whole adult population of the wards. The first part of the study deals with the statistical material, i.a. some defects in the scale by which intelligence is expressed, the advantages and disadvantages of the fact that test results were not available for the recruits rejected at the medical examination and the shape of the frequency distributions of the results per ward. The inter‐ward differences in intelligence are submitted to an analysis of variance and prove to be highly significant. The second part of the study gives an enumeration of the results obtained by correlating intelligence with various demographic and sociological factors. Very significant and positive correlations were found with income (average income assessed for income tax) and social status (measured by the percentage of unskilled workers among the occupationauy engaged population). The high correlation between intelligence and social status, in itself probably due to a process of social selection based on intelligence, may also be drawn upon to explain the results obtained as to political conviction. This conviction measured by the percentage of votes per ward as given to the various political parties, shows highly significant coefficients for the communists and the liberal party, negative in the former and positive in the latter case. For the labour party and the Roman‐Catholic party, two important groupings chiefly appealing to the middle classes, no significant correlations are found. Fertility in marriage, generally considered as having strong links with intelligence, gives rise to a hardly significant though negative correlation. This result, however, can considerably be improved, although still remaining rather low, by introducing the percentage of Roman‐Catholics among the population of the wards. This is in accordance with the well‐known fact that the fertility of the various religious groups in the Netherlands is strongly diverging, being highest among the Roman‐Catholics. A very high correlation again is obtained with the number of children begot before marriage as a percentage of the total number of first born (children born alive within J months after contracting marriage). Here probably the influence of changing social standards in the various wards makes itself felt. The last part of the article gives an interpretation of the results of the investigation. Although the significant correlations under certain reservations may be interpreted as being due to causal interrelationship, intelligence itself need not be a direct cause. A model may be conceived in which all characteristics together with intelligence are explained by a number of hereditary and environmental factors. Now assuming that intelligence as measured in this investigation is entirely free from environmental influences (there are some indications that this is not so) it can be interpreted as representing only one aspect of the hereditary complex. But even under this assumption many other factors may be involved in a complete explanation of the characteristics. A factor analysis which may perhaps partly clear up the complicated network of interrelations is under consideration.