ABSTRACT
This study demonstrates a post-hoc segmentation that is effective in creating distinct and robust segments of interest for researchers and practitioners in science communication. We use agglomerative hierarchical clustering to classify survey respondents based on answers to a short, five-item battery of questions drawn from variables that are frequently used in science communication. Resulting segments demonstrate strong differences in regards to demographics, issue-specific science knowledge, attention to media, sentiment toward media, and social networking site use. We assess the utility of the segments through a series of regression analyses in order to determine differences among segments with regards to intent to seek information about three science issues: fracking, nanotechnology, and synthetic biology.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Regulations requiring the disclosure of chemicals used in fracking went into effect in 2015 after the administration of this survey.