Search
On-Site Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Unit
Browse By Session Type
Search Tips
Change Preferences / Time Zone
Sign In
Bluesky
Threads
X (Twitter)
YouTube
This study examines the use of propagandistic language in recent state education department press releases. Using classifications from a BERT model trained on expert annotated articles, I examine how propaganda relates to time, spending per student, graduation rate and gubernatorial party affiliation. Republican leadership and spending per student were positively related to propagandistic classifications, and graduation rate was negatively related. Further, BERTopic models provide insight into how topics differ across time, propaganda classifications, and gubernatorial party affiliations. Interestingly, propagandistic articles reflected contentious issues, including ethnic studies, inclusion, civics education and parental rights in education. Non-propagandistic articles featured less controversial issues like awards, meal programs, and education quality. More broadly, I demonstrate the utility of NLP in research evaluating digital media.