The logic in ecological: II. The logic of design.

Abstract

The utility of ecological studies is considered in terms of the salience of their designs and is exemplified in four levels: obligate and apt; optional and apt; optional, not apt but convenient; and maladroit (neither obligate, apt, nor justifiable by convenience). Ecological studies are obligate when they are the only choice available, either because of the question asked (as in testing differences between groups and discovering group effects), or where there are "dependent happenings" (as in transactions involving more than one individual), or because individual data are lacking. Apt studies are logically appropriate; analysis and results are not extrapolated beyond necessity or without precautions. Obligate studies enforced by lack of individual data may be apt or less than apt. Optional ecological studies may be apt or, if less than apt, they may yet be convenient. Maladroit studies are neither obligate, apt, nor convenient. Each class of study is illustrated by examples ordered according to a standard design hierarchy.