Abstract
Let (X, Y, Z) be a trivariate statistical variable observed at individual level. We propose a three-term decomposition of covariance between variables X and Y conditionally on the effects induced by the existence of a variable Z. The three terms are called residual covariance, covariance lack of fit and covariance fit, respectively. Partial covariance, between X and Y after removing the linear effects of Z, σ Z (X, Y), is the sum of the first two terms while ecological covariance, between the two regression functions μ X (Z) and μ Y (Z), Cov(μ X (Z), μ Y (Z)), is the sum of the last two terms and, consequently, covariance lack of fit is the common additive term. Simple examples are given in two contexts: in ecological fallacy problems arising in linear modelling with aggregate level analysis contrasted with partial correlation step-wise procedures performed at individual level and in the special case of a two-level nested model. Previous basic decomposition is extended to a multivariate-multiple framework. Distinction between descriptive and stochastic approaches is not essential.
Similar content being viewed by others
References
Achen CH, Shilvely WP (1995) Cross-level inference. University of Chicago Press, Chicago
Faraway JJ (2005) Linear models with R. Chapman and Hall/CRC Press, Boca Raton
Freedman DA, Klein SP, Sacks J, Smyth CA, Everett CG (1991) Ecological regression and voting rights. Eval Rev 15: 673–711 (with discussion)
Freedman DA, Klein SP, Ostland M, Roberts MR (1998) Review of a solution to the ecological inference problem. J Am Stat Assoc 93:1518–1522 with discussion, vol 94, pp 352–357 (1999)
Guseo R (2006) Statistica, 3rd edn. Cedam, Padua
King G (1997) A solution to the ecological inference problem. Princeton University Press, Princeton
King G, Rosen O, Tanner MA (2004) Ecological inference: new methodological strategies. Cambridge University Press, Cambridge
Robinson WS (1950) Ecological correlations and the behavior of individuals. Am Soc Rev 15: 351–357
Schuessler AA (1999) Ecological inference. Proc Natl Acad Sci USA 96: 10578–10581
Author information
Authors and Affiliations
Corresponding author
Additional information
This research was partially supported by Fondazione Cassa di Risparmio di Padova e Rovigo, Progetto di Eccellenza 2007: “Innovation Diffusion Processes: Differential Models, Agent-Based Frameworks and Forecasting Methods”.
Rights and permissions
About this article
Cite this article
Guseo, R. Partial and ecological correlation: a common three-term covariance decomposition. Stat Methods Appl 19, 31–46 (2010). https://doi.org/10.1007/s10260-009-0117-0
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10260-009-0117-0