Vibes to Variables

A Methods Package for Students New to Research, Stats, and Code

Author

Alex P. Leith

Published

May 2026

Vibes to Variables

Third Edition (V2V)

This open educational resource teaches communication and media research methods as a complete arc, from the first spark of curiosity through a published, reproducible study. V2V is written for students who have never opened a stats program and have never written a research paper. It meets them where they are and walks them to a defensible study.

V2V is a methods package, not a textbook with bonuses. The book you are reading is one of several co-equal components:

The book (this site): chapter prose anchored to a real, public dataset.
V2V Hub: the course site that wraps this book with workbooks, cheat sheets, datasets, references, and instructor materials.
v2v R package: friction-reducing helpers that make chapter exercises one-line calls instead of multi-step setup.
Beyond Vibes: the graduate supplement, paired with MC 500, that produces a publishable white paper and conference poster.

How This Book Works

The book is organized into 14 chapters across five parts that mirror the phases of a research project:

Foundation (Chapters 1 to 3): Reframe research as disciplined storytelling, set up the workspace, ground epistemology and ethics.
Planning (Chapters 4 to 6): Literature review, theoretical lens, and a research prospectus that locks scope before any code.
Operationalization (Chapters 7 to 9): Structured listening, operationalization (NOIR), and first R contact through the codebook.
Execution (Chapters 10 to 12): Sampling and inter-coder reliability, data wrangling, visualization.
Inference and Publication (Chapters 13 to 14): Hypothesis testing with effect sizes, then a one-click reproducible portfolio rendered to GitHub Pages.

V2V 3rd Edition: in development

This third edition is a structural rewrite of the second edition: 22 chapters condensed to 14, R deferred to Chapter 9 (so the first eight chapters build conceptual scaffolding before any code), and every example anchored in the Twitch working corpus rather than four separate datasets. Chapters 1 through 6 are drafted; chapters 7 through 14 are stubs at this release and will land progressively through Summer and Fall 2026. The V2V 2nd Edition (22 chapters, music dataset) remains available at the _archive-v2/ folder in this repository as a transitional reference.

The Dataset

V2V 3rd Edition anchors all examples in the Twitch working corpus, a 1.6 GB PostgreSQL dump from a 2018 dissertation: two tables (chat_log for IRC messages, stream_log for Twitch API snapshots) covering nearly 1,700 channels across five and a half days of public broadcasting. The v2v R package ships row-sampled fixtures so students never touch the full dump:

library(v2v)
v2v::twitch_chat()    # sampled chat_log
v2v::twitch_streams() # sampled stream_log

The 2nd Edition’s unified_music dataset (Billboard + Spotify + Genius for 1,792 songs) remains available in the v2v R package as a secondary teaching corpus.

License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Author: Alex P. Leith, Southern Illinois University Edwardsville

Courses: MC 451 Social Media Analytics (undergraduate); MC 500 + MC 501 Research Methods (graduate, via the Beyond Vibes supplement)

--- title: "Welcome" --- # Vibes to Variables {.unnumbered} *Third Edition (V2V)* This open educational resource teaches communication and media research methods as a complete arc, from the first spark of curiosity through a published, reproducible study. V2V is written for students who have never opened a stats program and have never written a research paper. It meets them where they are and walks them to a defensible study. V2V is a methods package, not a textbook with bonuses. The book you are reading is one of several co-equal components: - **The book** (this site): chapter prose anchored to a real, public dataset. - **V2V Hub**: the course site that wraps this book with workbooks, cheat sheets, datasets, references, and instructor materials. - **`v2v` R package**: friction-reducing helpers that make chapter exercises one-line calls instead of multi-step setup. - **Beyond Vibes**: the graduate supplement, paired with MC 500, that produces a publishable white paper and conference poster. ## How This Book Works The book is organized into 14 chapters across five parts that mirror the phases of a research project: 1. **Foundation** (Chapters 1 to 3): Reframe research as disciplined storytelling, set up the workspace, ground epistemology and ethics. 2. **Planning** (Chapters 4 to 6): Literature review, theoretical lens, and a research prospectus that locks scope before any code. 3. **Operationalization** (Chapters 7 to 9): Structured listening, operationalization (NOIR), and first R contact through the codebook. 4. **Execution** (Chapters 10 to 12): Sampling and inter-coder reliability, data wrangling, visualization. 5. **Inference and Publication** (Chapters 13 to 14): Hypothesis testing with effect sizes, then a one-click reproducible portfolio rendered to GitHub Pages. ::: {.callout-note} ## V2V 3rd Edition: in development This third edition is a structural rewrite of the second edition: 22 chapters condensed to 14, R deferred to Chapter 9 (so the first eight chapters build conceptual scaffolding before any code), and every example anchored in the Twitch working corpus rather than four separate datasets. **Chapters 1 through 6 are drafted; chapters 7 through 14 are stubs at this release** and will land progressively through Summer and Fall 2026. The V2V 2nd Edition (22 chapters, music dataset) remains available at the `_archive-v2/` folder in this repository as a transitional reference. ::: ## The Dataset V2V 3rd Edition anchors all examples in the **Twitch working corpus**, a 1.6 GB PostgreSQL dump from a 2018 dissertation: two tables (`chat_log` for IRC messages, `stream_log` for Twitch API snapshots) covering nearly 1,700 channels across five and a half days of public broadcasting. The `v2v` R package ships row-sampled fixtures so students never touch the full dump: ```r library(v2v) v2v::twitch_chat() # sampled chat_log v2v::twitch_streams() # sampled stream_log ``` The 2nd Edition's `unified_music` dataset (Billboard + Spotify + Genius for 1,792 songs) remains available in the v2v R package as a secondary teaching corpus. ## License This work is licensed under a [Creative Commons Attribution 4.0 International License](https://creativecommons.org/licenses/by/4.0/). **Author:** Alex P. Leith, Southern Illinois University Edwardsville **Courses:** MC 451 Social Media Analytics (undergraduate); MC 500 + MC 501 Research Methods (graduate, via the *Beyond Vibes* supplement)