ABSTRACT
In psychology, data analysis proceeds according to a standard approach that is focussed on inferences about states of affairs obtained within a population of homogeneous human subjects. The information that is employed to infer about such states of affairs at the population level is of a particular type, namely interindividual variation (variation between subjects; individual differences). This approach underlies all standard statistical analysis techniques such as analysis of variance, regression analysis, path analysis, factor analysis, and multilevel and mixed modeling techniques. Whether the data are obtained in cross-sectional or longitudinal designs, the statistical analysis is always focussed on the structure of interindividual variation at the population level. Parameters and statistics of interest are estimated by pooling across subjects, which implies that these subjects are assumed to be homogeneous in all relevant respects. This is the hallmark of analysis of interindividual variation: the sums defining the estimators in statistical analysis are taken over different subjects randomly drawn from a population of presumably homogeneous subjects.
