RESUMO
In this chapter, we demonstrate the way certain common analytic approaches (e.g., polynomial curve modeling, repeated measures ANOVA, latent curve, and other factor models) create individual difference measures based on a common underlying model. After showing that these approaches require only means and covariance (or correlation) matrices to estimate regression coefficients based on a hypothesized model, we describe how to recast these models based on time-series related approaches focusing on single subject time series approaches (e.g., vector autoregressive approaches and P-technique factor models). We show how these latter methods create parameters based on models that can vary from individual-to-individual. We demonstrate differences for the factor model using real data examples.
