Abstract
This paper examines common modeling assumptions in biologically based sex research and critiques two recently published studies about homosexuality. The studies purported different conceptions of sexual attraction and different causal models and conclusions. Both studies employed large samples, complex statistical analyses, and assumptions about gender role behavior. Critical examination finds that gender role assumptions hinder explication of sexual orientation, and theories about Western homosexuality contribute little to a general science of sexual orientation. Mechanistic androgen models of sexual orientation overly simplify a multi-factorial, interactive developmental process. Researchers of sexual orientation should identify and discuss their modeling and causative assumptions, avoid Western sexual categorization, examine varied erotic attractions, and include cross-cultural comparisons.