MSNBC Transcripts

Collection of MSNBC show transcripts from 2008–2022.

Download

All data are available on Harvard Dataverse: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/UPJDE1

Available Data

API Scrape (May 2026) — Recommended

10,739 transcripts across 216 shows, May 2010 – October 2022.

Year	Transcripts
2010	43
2011	115
2012	233
2013	237
2014	217
2015	994
2016	906
2017	1,185
2018	1,467
2019	1,475
2020	1,288
2021	1,471
2022	1,108

Files:

• msnbc_transcripts_api_2010-2022.tar.gz — HTML transcript files (186MB)
• msnbc_transcripts_api_2010-2022_metadata.csv — metadata for all transcripts
• msnbc_shows_api.csv — show list with transcript counts

Legacy Scrape (2008–2014)

5,369 transcripts from the now-defunct http://www.nbcnews.com/id/3719710, scraped via archive.org.

Year	Transcripts
2008	76
2009	434
2010	752
2011	1,042
2012	1,164
2013	1,177
2014	724

Files (on Dataverse):

• Raw HTML files and final CSV available at: https://doi.org/10.7910/DVN/ND1TCV

Local files:

• data/legacy_2008-2014/legacy_2008-2014_links.csv — list of all transcript URLs
• data/legacy_2008-2014/legacy_2008-2014_links_with_metadata.txt — URLs with show names and dates

Notes:

• Scripts from 2014
• Some transcripts had date typos (e.g., 'Thusday', 'Februrary') causing parsing failures

Earlier Scrapes

• 2003–2014 scrape: 16k transcripts from an earlier collection
• 2025 HTML scrape: msnbc_transcripts_2022.csv.gz — transcripts from 2020–2025 scraped from listing pages

Data Format

Metadata CSV (msnbc_transcripts_api_2010-2022_metadata.csv)

Column	Description
id	Unique transcript ID
date	Publication date (ISO 8601)
title	Transcript title
url	Original URL
slug	URL slug
guests	Guest names
show_ids	Associated show IDs
modified	Last modified date

Shows CSV (msnbc_shows_api.csv)

Column	Description
id	Show ID
name	Show name
slug	URL slug
count	Number of transcripts

HTML Files

Each transcript is saved as an HTML file named {id}.html containing the full transcript text.

Scripts

For reproducibility or extending the dataset:

• WordPress API Scraper — REST API scraper (recommended)
• HTML Scraper — scrapes transcript listing pages
• Quick Peek — preview data
• Upload to Dataverse

Legacy scripts (2008–2014):

• Legacy Crawl — crawls archive.org for transcript links
• Legacy Extract — extracts transcripts from HTML files