Framed by the Feed: Social Media’s Influence on ADEMIC RESEARCH POSTEREnvironmental TEMPLATE Sentiment Subtitle for Academic Research Poster (48x36 inches)
Benjamin Izaguirre Cortez Your names and the names of the people who contributed to this presentation Department of Geology, Environment, and Sustainability, Research Mentors: Dr. Sasha Pesci
Introduction
• In the digital age, social media has emerged as a powerful platform for shaping public opinion.
Research questions • How does emotional framing in environmental videos influence user sentiment? • How do major events such as natural disasters or sustainability campaigns affect audience sentiment over time? • Which framing styles are most influential in shaping viewer sentiment?
Data Analysis
Key findings
• • •
Sentiment Score normalized using: Z = (x - μ) / σ • μ = 0.1049 (mean sentiment) • σ = 0.6195 (std deviation) Outliers: • Strong Positive (Z > 1.5): 4.86% • Strong Negative (Z < -1.5): 6.41%
• Positive framing generated higher sentiment and stability.
Platform: YouTube (Spring 2025) Sample: 3,991 comments across 27 videos Analysis: VADER Sentiment + Z-score normalization • Event Types: Natural Disasters, Awareness Campaigns, Control • Framing Styles: Positive (Hopeful), Crisis (Urgent), Informative (Neutral) • Tools: Python, Excel, OpenPyXL, VSCode
• Previous research suggests that framing strategies on social media can influence public sentiment, emotional engagement, and user behavior (Valenzuela et a., 2017; Zuniga, 2022; Guenter et al., 2020)
• Employing a mixed-methods approach, the research analyzes viewer comments collected from a diverse range of YouTube creators to understand how video content, emotional appeal, and audience engagement collectively influence sentiment on sustainability topics.
Methods
Figure 1. Count of Sentiment Distribution per Category
Figure 3. Channel A2 (Satirical / Activist) Sentiment Distribution Over Time
2000
• Crisis content triggered high engagement but polarized response.
• Educational creators with visual or hopeful tones outperformed neutral ones. • Sentiment was less polarized for “Climate Change” topics compared to other specific disaster events (e.g., wildfires).
1800
• Event-timed content (e.g., Earth Day) amplified sentiment shifts.
Figure A: neque dignissim, and in aliquet nisl et umis.
1600 1400 1200
1000
Conclusions
800 600
400 200 0 Negative
Neutral
Positive
Figure 4. Avg Sentiment Z-Score per Channel
Figure 2. Avg Sentiment Z-Score per Event Category Natural Disasters
Results
Row Labels
Wildfires
• Content with positive framing had the highest average sentiment.
Hurricane
• Crisis or ideological framing reflected more polarized sentiment.
Green Solutions Conservation
• Some sentiment spikes and troughs were correlated with event timing.
Climate Change Activism -0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
Average of Zscore
Channel B2 Channel A2 Channel A1
-0.146793262
Channel C1
-0.226772312
Channel C2 Channel B3 Channel B1
-0.091726482
Channel D2 Channel D1 0.4 Grand Total
-0.213290332
• Framing matters: Tone affects audience emotion. • Positive/hopeful content supports engagement without eco-anxiety. • Findings support more emotionally effective environmental messaging online.
References
0.085224824 0.179039761
Acknowledgements Special thanks to Dr. Sasha Pesci, the Department of GES, and the National Center for Suburban Studies for their continued support in this research project.
0.086868906 0.075643363
0.04846369 -0.03227937 Department of Geography
buffalo.edu