Pub. date: 2008 | Online Pub. Date: November 27, 2007 | DOI: 10.4135/9781412953948 | Print ISBN: 9781412928168 | Online ISBN: 9781412953948| Publisher:SAGE Publications, Inc.About this encyclopedia
John J. Hsieh
Relationships observed for groups do not necessarily hold for individuals, and vice versa. The ecological fallacy is a fallacy in ecological studies that may arise when an investigator makes an inference about an individual based on aggregate data for a group. In ecological studies, we assess the relation between exposure rates and event rates at group level because we know only marginal distributions of exposure (risk factor) and outcome event and not their joint distributions. Researchers have made unwarranted inferences from the association between exposure to risk factor and outcome event among groups (ecological data) to association among individuals within each group without accounting for the possible ecological bias. Aggregating data loses information. The ecological fallacy may arise because the process of aggregating data may conceal the variations that are not visible at the larger aggregate level (see explanation in Example 2 below). Statistically, a correlation tends to be Robinson ...