In its simplest terms, Spearman’s Law of Diminishing Returns (SLODR) is the notion that at higher levels of ability, a general intelligence factor is a less meaningful determinant of cognitive ability than at lower levels... [ view full abstract ]
In its simplest terms, Spearman’s Law of Diminishing Returns (SLODR) is the notion that at higher levels of ability, a general intelligence factor is a less meaningful determinant of cognitive ability than at lower levels of ability. Since Detterman and Daniel (1989) rediscovered the phenomenon, research on SLODR has generally found that the influence of g is differentiated by cognitive ability. However, no research has previously explored whether this phenomenon operates differently across racial/ethnic groups or across the sexes.
Traditionally, research on SLODR has been conducted by comparing two subtest intercorrelation matrices for a high- and low-ability group on a given intelligence measure; the high-ability groups tend to have lower average intercorrelations between intelligence tasks than do the lower-ability groups. Other methods, such as principal component analysis and exploratory factor analysis have been used to accomplish the same goal: comparing high- and low-ability groups.
For the present study, confirmatory factor analytic models will be created that compare high- and low-ability groups across race and sex groups. Data will come from the nationally representative standardization sample of the Differential Ability Scales—Second Edition. The total standardization sample is made up of 2,952 individuals; our study, however, will only include children and adolescents between the ages of 7:0 and 17:11. The sample will be age-limited as these children were administered a common battery of tests. Because the selected sample will include a much larger number of Caucasian participants, a random sample of Caucasian participants will be selected that is equivalent in size to the next largest subsample for the ethnic group comparisons. For the sex differences question, all males and females will be compared.
The research is still in progress. However, we expect these findings will show no differences in how SLODR operates between racial/ethnic groups or sex groups. Nevertheless, this research is still needed, especially considering the use of intelligence testing to make determinations of special education eligibility and gifted program placement. The presence of SLODR indicates that the current theoretical model of intelligence may not be the most practically useful. For example, SLODR may help to explain why overall g is an inherently limited way of assessing people of above-average ability and making determinations about giftedness curricula. Furthermore, SLODR would suggest that, for higher-ability individuals, g would be less predictive of future job performance than would domain-specific broad ability scores. Because the implications of SLODR are so compelling, it therefore follows that SLODR should be understood as fully and completely as possible. Despite the amount of research into SLODR, ethnic similarities and differences, and sex similarities and differences, no study has attempted to understand the interrelationship of these topics.
