Intelligence differentiation in adult samples

Spearman's “Law of Diminishing Returns” tested with two methods

“…When consulting the more valid studies (e.g., larger studies controlling for differences in variance across groups) the actual size of the differences in g saturation, as measured by average inter-correlation, only amounts to a small effect size (estimated mean; r c 0.1) across an ability span of 1 1 / 2-2SD. This is in concordance with some of the previously mentioned studies that find a difference of approximately 0.1-0.2 (e.g., Abad et al, 2003;Jensen, 2003;Legree et al, 1996) whereas others (e.g., Abad et al, 2003;Deary et al, 1996) find a difference of below 0.05, just to mention a few of the newer and more valid studies. With this effect size there is by no means certainty of the true effect of SLODR (ability), since in order to detect an effect size of 0.1 with a power of 80%, one needs a sample of at least 1573 in each group (Cohen, 1992).…”

“…When consulting the literature the majority of studies support SLODR (ability) (e.g., Abad, Colom, Juan-Espinosa, & Garcia, 2003;Baumeister & Bartlett, 1962;Birren, 1952;Carlstedt, 2001;Coyle, 2002;Deary et al, 1996;Der & Deary, 2002;Detterman & Daniels, 1989;Detterman et al, 1992;Evans, 1999;Filella, 1960;Jäger, 1964;Jäger & Todt, 1964;Jensen, 2003;Kane, 2000;Legree, Pifer, & Grafton, 1996;Lienert, 1961Lienert, , 1964Lienert & Faber, 1963;Lynn, 1992;Lynn & Cooper, 1993, 1994Mitchell, 1956;Maxwell, 1972;Nesselroade & Thompson, 1995;Reinert, Baltes, & Schmidt, 1965;Segel, 1948;Spearman, 1927;Wewetzer, 1958) but some studies either find no difference in g saturation or find the opposite pattern (Amelang & Langer, 1968;Bloom et al, 1988;DeVoss, 1926;Eyferth, 1963;Facon, 2004;Fogarty & Stankov, 1995;Hartmann & Teasdale, 2004, accepted for publication; Roesslein, 1953). However, not all studies are equally valid and it is therefore necessary to look at the individual studies in order to investigate whether the specific study controls for potential confounders.…”

“…This approach suggests that since an ordinary IQ test is designed for the average subject, then High ability subjects will find these tests too easy and hence the positive manifold will diminish thereby decreasing the g saturation. Still others suggest that the differentiation as a function of ability is probably due to increase in education (e.g., Colom et al, 2002;Abad et al, 2003). Increased education goes with increased ability, and more education will result in a more differentiated ability structure due to the increased specific knowledge obtained education.…”

“…However, by ensuring the same discriminatory items along the entire distribution, identical reliability can be ensured for all subjects. This can be of importance since differences in reliability across ability groups can affect the results (e.g., Abad et al, 2003;Burt, 1954;Deary et al, 1996;Deary & Pagliari, 1991;Detterman, 1991;Fogarty & Stankov, 1995;Graham, 1953;Jensen, 2003) by making the groups with less reliability artificially obtaining lower g saturation. Despite the fact that studies do not support differential reliability across ability groups (Deary et al, 1996;Fogarty & Stankov, 1995;Juan-Espinosa et al, 2002; and partially by Abad et al, 2003; Hartmann & Teasdale, accepted for publication), having a test format that a priori potential ensures the highest and most uniform reliability across groups is an advantage in the study of SLODR (ability) where many confounding variables can affect the results.…”

A diferenciação cognitiva na infância: um estudo de perfis cognitivos aos 5, 7 e 9 anos

“…Esse padrão mais diferenciado das crianças com QI elevado parece corroborar a teoria SLODR (Spearman, 1927), também replicada em estudos mais recentes (Coyle & Rindermann, 2013;Reynolds & Keith, 2007;Reynolds, Keith, & Beretvas, 2010;Tucker-Drob, 2009). A razão dessa relação inversa tem tido várias explicações, principalmente uma melhor eficiência dos processos cognitivos centrais e do sistema de processamento de informação (Detterman, 1991(Detterman, , 1999 ou uma maior apropriação das experiências escolares (Abad, Colom, Juan-Espinosa, & García, 2003) pelos indivíduos com QI elevado. Finalmente, e ao contrário de outros estudos (Brito, Almeida, Ferreira, & Guisande, 2011;Johnson & Bouchard, 2005), não se observou uma diferenciação cognitiva em função dos conteúdos das tarefas.…”