European and American WAIS III norms: Cross-national differences in performance subtest scores Eka Roivainen ⁎ Oulu Deaconess Institute, PL 365, 90101 Oulu, Finland article info abstract Article history: Received 23 June 2009 Received in revised form 5 October 2009 Accepted 5 October 2009 Available online 1 November 2009 For this study, European WAIS III performance subtest norms were compared to the original US norms. When European WAIS III raw scores were scored using US norms, the resulting perceptual organization index (POI) means were significantly higher than the processing speed index (PSI) means. The POI/PSI difference is roughly 5–10 points for the German (n = 1897), Spanish (n = 1369), French (n = 1104), British (n = 370) and Finnish (n = 446) WAIS III stan- dardization samples. The POI mean was higher than the PSI mean in all age groups in each European sample. It is concluded that cross-national differences in any one factor (such as general intelligence) are unlikely to cause such effects. It is hypothesized that cultural factors, differences in education and test-taking attitudes, underlie the difference between US and European test profiles. The results of the study suggest that national test norms may have poor cross-national validity even when the effects of age, sex and education are controlled for. © 2009 Elsevier Inc. All rights reserved. Keywords: Intelligence Norms Cross-national WAIS III 1. Introduction The Wechsler adult intelligent scale (WAIS) III was published in the United States in 1997 (Wechsler, 1997) and has since been translated and adapted for use in several non- English-speaking countries. The Spanish version was pub- lished with national norms in 1999 (Wechsler, 1999a), the French version in 2000 (Wechsler, 2000) and the German version in 2006 (Wechsler, 2006). In smaller countries, such as Sweden (Wechsler, 2003) and Denmark (Wechsler, 2005a), the test instructions and relevant items were translated, but national norms were not created. For the British version (Wechsler, 1999b), test instructions and items were “angli- cized” but American norms are used for scoring the test. In Britain, a sample of 332 persons was drawn to test the “anglicized” WAIS III (Wycherley, Benjamin, Crawford & Mock- ler, 1999). The IQ and index means as well as six subtest means were significantly higher than in the original American sam- ple. This result was interpreted in terms of a small IQ drift, “the Flynn effect,” (Flynn, 1998), and a sampling bias leading to the collection of a more capable group than intended. In a sub- sequent study (Wycherley, Lavender, Holttum, Crawford & Mockler, 2005), 125 participants were removed from the orig- inal sample and 163 new participants were added to create a new sample that would more reliably represent the UK pop- ulation for educational and occupational level. However, also in this augmented sample, the IQ means (shown in Table 1) and many subtest means were significantly higher than in the US sample. The subtest means varied from 10.0 (digit symbol coding) to 11.3 (block design) (Table 3). Only the processing speed index (PSI) did not differ from the US mean. Based on the IQ profiles of the sample, it was hypothesized that a sam- pling bias caused by volunteer participants, the overrepresen- tation of intellectually able but poorly educated participants in the older age groups, caused the observed difference. A statistical procedure based on the assumption that the PSI reflected the “unbiased intellectual level of the sample” was devised to correct the mean IQ and subtest scores. Finally, it was concluded that US norms can be safely used with the UK population. In a recent unpublished study (Roivainen, in press), the Finnish WAIS III (Wechsler, 2005b) performance subtest norms based on the Finnish standardization study (n = 446) were Intelligence 38 (2010) 187–192 ⁎ Tel.: +358 443535625. E-mail address: eka.roivainen@odl.fi. 0160-2896/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.intell.2009.10.001 Contents lists available at ScienceDirect Intelligence