TikTok Slang Diffusion

ISEF Category: Behavioral and Social Sciences

Subcategory: Sociology and Anthropology  ·  Difficulty: Intermediate  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

A slang word can jump from a small group to huge audiences in days. TikTok leaves public metadata that lets you track that jump. You can ask whether one big creator acts like a megaphone, or whether tight communities spread the term first. That turns a trend into a testable network question.

What Is It?

This project asks how a new word or phrase moves online. In linguistics, a lexical item is just a word or phrase. Diffusion means spread over time. You track where the term appears first, then map who follows whom, who reuses it, and which communities pick it up next. Think of it like a rumor in a school, where one popular student might spread it fast, or friend groups might pass it along inside the group before it crosses over.

TikTok public metadata can give you clues like timestamps, creator IDs, hashtags, sounds, and engagement signals. You can use those clues to build a network graph, which is a map of connections. Then you compare hub-centered spread, where a few large accounts matter most, with cluster-centered spread, where smaller groups pass the term along inside local networks. The key idea is simple, you are not just asking who used the slang, you are asking how it moved.

Why This Is a Good Topic

This topic works well because you can measure real spread, not just guess about trends. It connects language change, online culture, and the way platform networks shape what people see. You can learn network mapping, basic statistics, and how to turn messy social data into a clear claim.

Research Questions

• How does the first appearance of a slang term differ between influencer-heavy networks and community-cluster networks?
• What is the effect of creator follower count on how quickly a new lexical item reaches later posts?
• Does the density of a TikTok community predict how fast slang spreads inside that cluster?
• To what extent do reposts, duets, or stitched replies accelerate lexical diffusion across clusters?
• Which network pattern better predicts whether a slang term crosses into new communities?
• How does the time lag between first use and broader reuse change when the term starts with a hub account?

Basic Materials

• Laptop with internet access and enough storage for CSV files.
• Spreadsheet software like Google Sheets or Excel.
• Python installed with pandas, NetworkX, and matplotlib.
• Jupyter Notebook or a similar code notebook.
• A text document for coding slang examples and notes.
• Cloud backup or an external drive for data files.

Advanced Materials

• Access to a university-approved social media data archive or research API.
• A workstation or cluster that can handle larger graph files.
• Python with network analysis and statistical libraries.
• Gephi for graph visualization.
• R or Stata for regression and survival analysis.
• Secure storage for de-identified data and audit logs.

Software & Tools

• Python: Cleans metadata, builds graphs, and measures diffusion timing.
• Jupyter Notebook: Keeps code, notes, and plots together.
• NetworkX: Builds network graphs and computes centrality, clustering, and reach.
• Gephi: Visualizes hub and cluster spread patterns.
• pandas: Organizes posts, creators, timestamps, and coding labels.

Experiment Steps

1. Define one slang term or a small set of related terms, then set clear rules for what counts as a use.
2. Map the network around each post by deciding which metadata fields stand in for creator links, community ties, or exposure paths.
3. Build a timeline that records the first appearance, repeat use, and cross-community jumps for the term.
4. Decide which comparison tests hub-centered spread against cluster-centered spread, and name the network metrics you will compare.
5. Plan controls for account size, topic overlap, and time window so you do not confuse popularity with diffusion.
6. Choose one output that turns the results into a clear claim, such as a graph, table, or diffusion curve.

Common Pitfalls

• Counting likes or views as proof of diffusion, which confuses attention with actual reuse.
• Mixing post-level and creator-level records, which double counts the same slang use.
• Ignoring the first timestamp for each term, which breaks spread order and makes direction unclear.
• Comparing big influencers with small communities without normalizing for audience size, which makes hub effects look larger than they are.
• Using inconsistent search terms or hashtag filters, which misses alternate spellings, abbreviations, and deleted posts.

What Makes This Competitive

A stronger version of this project would test several slang items, not just one, and compare them with a clear null model. You could separate creator size, community density, and time order, then see which factor still matters after the others are controlled. A strong entry would also use a careful network metric, not just a pretty graph. That kind of analysis shows you can make a real claim about diffusion, not just describe a trend.

Project Variations

• Compare slang spread in niche fandom communities versus broad entertainment communities to see whether group structure changes diffusion speed.
• Track one slang term across duet, stitch, and comment networks to test whether reply behavior changes who adopts the term next.
• Compare slang terms started by large creators with terms started by mid-size creators to see whether hub size changes cross-community reach.

Learn More

• Pew Research Center: Find free reports on teen social media use and online behavior on the Pew Research Center site.
• TikTok Research API documentation: Read the public documentation for available metadata fields, access rules, and search limits.
• MIT OpenCourseWare: Search for free network science or social network analysis lectures and assignments.
• Google Scholar: Search for review articles on lexical diffusion, sociolinguistics, and online community networks.
• Language in Society: Search for peer-reviewed articles on slang change and social meaning through Google Scholar or your school library.