Employing Multivariate Logistic Regression Analysis to Better Understand the Relationship between Individuals’ Anthropogenic Climate Change Acceptance and Belief in Anti-Climate Change Dissenter Messages
Date of Award
Doctor of Philosophy
Science Education, Mallinson Institute
Dr. Heather Petcovic
Dr. David Rudge
Dr. Sheldon Turner
Organized climate change dissention groups spend a considerable amount of time and energy developing messages designed to convince the public that anthropogenic climate change is neither a reality nor a threat. These messages work against the efforts of climate educators and can be divided into two categories; messages that provide alternative explanations for warming, or messages that attack the work of scientists studying anthropogenic climate change. There has been a lack of research regarding any correlation between individuals’ agreement with these messages and their rejection of anthropogenic climate change. Establishing a correlation would be an indication that educators should take steps to inoculate individuals from dissenter messages.
This dissertation project answers this broad question via an analysis of responses to the anthropogenic climate change dissenter inventory (ACCDI). This survey tool measures individuals’ agreement with dissenter messages across six factors: naïve scientific statements which assert no connection between atmospheric carbon dioxide and climate, anti-science statements which attack the credibility of climate scientists, sophisticated scientific statements which imply warming is not anthropogenic, arguments that assert recent changes are natural or out of our control, arguments that imply current warming is simply part of a larger cycle, and statements that highlight benefits of a warmer climate.
Amazon’s Mechanical Turk was used to recruit 689 participants. Participants’ responses on the ACCDI were subjected to multivariate logistic regression analysis. Results indicate that agreement with dissenter messages is a predictor of dissent among the study population. Particularly strong predictors are messages that attribute recent warming to natural climate cycles. Agreement with messages that describe anthropogenic climate change as beneficial, and agreement with naïve messages were not predictors of dissent. Other predictors include an individual’s preferred news network, and political ideology. The results of this dissertation strongly suggest that effective anthropogenic climate change instruction should include time dedicated to inoculating students from misinformation and discussion exploring why groups or individuals would spread this misinformation. This could be accomplished through misconception driven instruction where students discuss the flaws of logic within misconceptions to highlight why they are misconceptions. Teaching students the science behind, and the societal implications of, anthropogenic climate change policy may lead to a more climate literate public.
Restricted to Campus until
Bentley, Andrew Phillip Keller, "Employing Multivariate Logistic Regression Analysis to Better Understand the Relationship between Individuals’ Anthropogenic Climate Change Acceptance and Belief in Anti-Climate Change Dissenter Messages" (2017). Dissertations. 3174.