Psychometrics variance

Psychometric procedures like Spearman—Brown provide psychological windows on human variation akin to the microscope in biology and the telescope in astronomy. With an average inter-item correlation of 0.20, a mere 40 item scale can generate a composite whose common (reliable) variance is 91%. This is how psychometrics distils dimensions of common variance for submission to construct validation procedures. 

If brain function measures that correlate with ability test scores occur at different levels of description then it is of interest to discover whether constructs from more than one level share variance that relates to ability test scores. Three examples of this approach follow. All three discuss ways in which the correlation between inspection times and psychometric intelligence may be investigated further. 

The effort to understand those brain functions that account for variance in psychometric intelligence produced a set of replicated correlations with indices at different levels of explanation. Future work will add to these associations and should devote more effort to explaining these correlations. A part of this explanatory toil must come from linking individual differences at different levels of description. Some contributions toward that end were offered, including statistical modelling, pharmacological intervention, functional brain imaging and event related potential studies. 

Prediction 1. Factor analysis of reliable fitness measures obtained from a large representative sample of individuals from any wild species will reveal a general fitness factor, an JF factor, analogous to the g factor in psychometrics. This f factor will account for a substantial proportion of between-individual variance in the fitness measures, and will prove moderately heritable under normal ranges of environmental conditions. 

Miller If one is aware of certain aspects of evolutionary theory, such as mutation—selection balance and strategic polymorphisms, they provide interesting competing accounts of why is so heritable and variable. Many of the measures of fitness and variance in animal species that have been developed provide interesting analogues of psychometric tests, and they might be mutually illuminating. They have been in this symposium. 

Our data were submitted to descriptive analysis in terms of mean values, range and frequency distributions. Since the distributions of lead exposure parameters were skewed, a log transformation of values was applied. Pearson s correlation coefficients were calculated to evaluate the association between neuropsychological impairments and lead exposure. The level of significance assumed was p < 0.05 (one-tailed). To evaluate the influence of potential confounding variables on measured parameters, analysis of variance (ANOVA) was performed. In addition, we estimated the variance of psychometric tests explained by covariates (ALA-D, PbB, PbH, PbT) after regressing out the effects of demographic variables (ANCOVA). 

We estimated the variance of psychometric tests explained by covariates (ALA-D, PbB, PbH, PbT) after regressing out the effects of demographic variables. 

---