Students’ Interpretation of Uncertainty in Health News Articles

• In Event: Public Understanding of Science: The Educational Challenges of Scientific “Uncertainty”

Abstract

Objectives and Purposes. The purpose of this study was to understand how students’ interpret uncertainty in authentic Web news articles presenting studies about health issues.

Perspective(s) or theoretical framework. People frequently read new articles on the web to come to learn about science and health issues that affect their lives. Despite the typically correlational nature of studies using human participants, readers often interpret such studies as more causal, and thus less uncertain, in nature. This tendency is reflected in the practice of editors who often make a story more eye-catching by modifying the title (headline) and first paragraph (lead).

Methods. In this study, we tested whether students are inappropriately influenced in this direction by the title of an article. Undergraduates read news stories about health issues downloaded from the web. Each article reported on the results of a study and the data was always clearly correlational. We manipulated whether the headline and lead were inconsistent (causal conclusion) or consistent (correlational conclusion) with the findings of the study reported in the article. To ensure authenticity, we located articles from the Web that presented a causal title and lead for the correlational study. Students were instructed to read the news articles “for comprehension so you can give a friend advice about these health topics”. Then, from memory, they wrote a 2-3 sentence summary of the article. Finally, they evaluated their perception of the reporter (i.e., Competency, Knowledge, Trustworthiness, Intelligence, Scientific, & Careful), the quality of the article (i.e., Sound, Well-reasoned, Logical, Strong) and persuasiveness on 7 point scales. We included several other dimensions to disguise the target of study.

Data sources. We analyzed students’ summaries and recall in terms of causal and correlational relations. We also collapsed the ratings for perception of the reporter and the quality of the article to form 2 combined ratings. Finally, we examined their detection of a conflict between the reporter’s title and the support provided by the study.

Results. We found a very high base rate of summarizing correlational findings as causal (M = 47%). This tendency of a causal summary was higher when the title made a causal statement (M = 57%) compared to when the title made a correlational statement (M = 37%). Thus, as expected the headline/lead framed students’ interpretation of the study that they read. Contrary to our expectations, using a less conclusive correlational title did not lead to a more negative perception of the reporter or message. We think this is due to the fact that most participants did not detect the inconsistency between title and content.

Scientific significance. Many students appear to interpret any scientific study as providing evidence for a causal relationship and thus more certain than actually justified. We will discuss the complex processes involved in understanding uncertainty in reading health and science news stories, such as how readers ascertain whether a title is causal or correlational, how they understand descriptions of studies and whether they realize that different types of studies support different claims about scientific certainty.

Authors

• Anne Britt, Northern Illinois University

• Amanda Marie Durik, Northern Illinois University

• Brent Steffens, Northern Illinois University

• Kayley Bloss, Northern Illinois University

• John Baker, Northern Illinois University