Abstract (eng)
In the field of intelligence, the phenomenon of rising intelligence test performance over generations, that has become well known as the “Flynn effect”, is now widely researched. These increasing intelligence test performances are characterized by an incline of three to five IQ points per decade, with variations between countries and domains, which have been observable since the 1900s (Flynn, 1984). Interestingly, a stagnation and even reversal of this effect has been noticed in the last 30 years, especially affecting Scandinavian countries (Teasdale & Owen, 2005). What factors might cause the Flynn effect and its potential reversal has been examined by several researchers, nevertheless, a comprehensive explanation for a country- and domain specific Flynn effect has yet to be found. In the current study, the influence of test norm changes on different intelligence domains has been investigated, specifically focusing on the hypothesis that item obsolescence might be a potential factor contributing to the observed generational IQ changes. For this matter, a total of 100 participants (54f; M = 26.3 years, SD = 7.9) were administered three subscales of the Hamburg Wechsler Intelligence Test for Adults (HAWIE; Hardesty & Lauber, 1956) as well as its revision, the Hamburg Wechsler Intelligence Test for Adults Revised (HAWIE-R; Tewes, 1991). A two-way repeated-measures ANOVA showed significant differences between performances in the original and the revised test versions. Further analysis revealed significantly better performances of participants in the original test version for items in the “Arithmetic” subscale, thus suggesting a positive Flynn effect in the domain of fluid intelligence. In the “Vocabulary” subscale better performance in the revised test has been observed, indicating a negative Flynn effect in crystallized intelligence, meaning that ability seemingly decreased over time. These changes in performance were assessed by specifically looking at the item scores of the items that were changed from the original to the revised test version in order to determine if the replacement of items is of considerable importance regarding generational changes of intelligence test performance. The significantly higher performance in the revised test versions on crystallized items points towards the effects of item obsolescence on IQ changes, masking crystallized IQ gains and thus exaggerating domain specific IQ differences in the Flynn effect.