Ecological Inference and the Ecological Fallacy: An Examination of the Use of Aggregate Geographic Data in Public Health Communication Planning

• Activity Number: 188
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Health Policy Statistics Section
• Author(s): William Pollard*+
• Keywords: Public health communication ; Health communication surveys

Abstract

Data for understanding at-risk target audiences are essential for developing effective communication. One type of data for health and communication-relevant socioeconomic and behavioral variables is that which is aggregated by census and postal geographies in existing public and commercial data bases. Such data are more readily accessible than individual-level data because of confidentiality issues. Associations between health variables and communication-relevant variables can guide communication planning. Such use of aggregate data to draw conclusions about relationships among individual-level variables is known as ecological inference. However because of information loss in aggregation this can lead to erroneous conclusions known as ecological fallacies. This issue is seldom discussed in the communication planning literature. The purpose of this presentation is to outline the nature of this problem and illustrate it by contrasting results obtained from aggregate and individual level data sets. Key literature is highlighted, and implications for communication planning are discussed.