PARENT SESSION
Symposium 9: Structural Equations, Path Analysis, and Other Causal Models
Organized by: S Scheiner
Tuesday, August 5. 1:30 PM to 5:00 PM, SITCC Chatham Ballroom C.

Path analysis and measuring natural selection.

Scheiner, Samuel^*,1, ¹ National Science Foundation, Arlington, VA

ABSTRACT- Crucial to understanding the process of natural selection is characterizing phenotypic selection. The traditional method uses multiple regression to dissect total selection into portions due to direct selection and indirect selection through other traits. However, such an analysis can mis-identify the causal components of phenotypic selection because the wrong causal model is being specified. Path analysis allows for more complex causal structures to be analysed. Path analysis also address two long-standing problems with selection analyses: the problem of missing traits and environmental bias. I show how path analysis can result in a different understanding of patterns of selection, and it strengths relative to a multiple regression approach. I also discuss available software that makes such analyses as easy to do as multiple regression.

Key words: natural selection, path analysis, structured equation modeling, multiple regression