codex dataset card

> generated by Codex_

Dataset Card for xjmr_260511

Dataset Summary

A structured crawl of public forum threads from Econ Job Rumors, a discussion site centered on economics academia, the economics job market, research, finance, political economy, and substantial off-topic discussion. The dataset contains 910,133 thread records and 8,863,628 extracted posts. Each row represents one (forum_slug, thread_slug) listing record and preserves thread metadata, forum/source-page metadata, snapshot vote and view counts, relative age/freshness strings, crawl timestamps, and one flat posts list for the thread.

The crawl targets public pages visible without authentication. Text bodies were extracted from topic HTML into normalized plain text.

---

Dataset Details

Dataset Sources

• Source Website: https://www.econjobrumors.com
• Source Pages: Forum listing pages and topic pages under https://www.econjobrumors.com/forum/... and https://www.econjobrumors.com/topic/...

---

Uses

Direct Use

Suitable for:

• Academic research into anonymous forum dynamics and rumor communities
• Computational social science research on economics academia and job-market discourse
• Linguistic analysis of informal, adversarial, anonymous, and professional-community web discourse
• Studying thread activity, post volume, view counts, and good/no-good voting patterns
• Information retrieval and semantic search over public Econ Job Rumors threads
• Moderation, toxicity, and safety research, with appropriate ethical review

Out-of-Scope Use

Not suitable for:

• Making decisions about named or identifiable people, departments, universities, journals, or employers
• Treating claims in posts as verified facts
• Harassment, doxxing, deanonymization, or targeted profiling of forum participants
• Inferring stable user identities from author_handle values

---

Dataset Structure

Schema

Column	Type	Description
thread_slug	string	Topic slug parsed from the thread URL
thread_url	string	Resolved thread URL used while crawling
canonical_thread_url	string	Canonical topic URL for the thread
title	string	Thread title parsed from the topic page, falling back to the listing title
listing_title	string	Thread title shown on the forum listing page
forum_section	string	High-level forum section, such as Economics, Job Market, or Off Topic
forum_name	string	Human-readable forum name
forum_slug	string	Forum slug parsed from the forum URL
forum_url	string	Canonical forum URL
source_page	int64	Forum listing page number from which the thread was discovered
source_page_url	string	Forum listing page URL from which the thread was discovered
listed_post_count	int64	Post count shown on the forum listing page
thread_post_count	int64	Post count parsed from the topic page metadata, if present
view_count	int64	View count shown on the forum listing page at crawl time
listed_votes_good	int64	Good votes shown on the forum listing page at crawl time
listed_votes_no_good	int64	No-good votes shown on the forum listing page at crawl time
thread_votes_good	int64	Good votes parsed from the topic page at crawl time
thread_votes_no_good	int64	No-good votes parsed from the topic page at crawl time
freshness_text	string	Relative freshness text shown on the forum listing page
thread_age	string	Relative age text parsed from the topic page metadata, if present
latest_post_url	string	URL for the latest post known from listing or topic metadata
latest_post_id	int64	Numeric latest-post ID parsed from latest_post_url
thread_pages_crawled	int64	Number of topic pages crawled for the thread
post_count_extracted	int64	Number of posts extracted into the posts list
crawled_at	string	ISO-8601 timestamp when the thread record was written
crawl_error	string or null	Crawl/fetch error recorded for the thread, if any
posts	list	Flat list of posts extracted from the thread

Post Struct

Field	Type	Description
thread_slug	string	Topic slug for the parent thread
post_id	int64	Numeric post identifier parsed from the post anchor
post_url	string	Direct URL to the post anchor
position	int64	1-based post position within the extracted thread
author_role	string	Role or display label shown in the post author block
author_handle	string	Handle text shown by the site for the post; may be pseudonymous or site-assigned
author_extra	string	Additional author metadata shown by the site, if present
posted_at_relative	string	Relative post age text shown by the site, such as 2 years ago
good_votes	int64	Good votes for the post at crawl time
no_good_votes	int64	No-good votes for the post at crawl time
body	string	Plain text post body extracted from HTML, with whitespace normalized

Statistics

Metric	Value
Thread rows	910,133
Unique canonical thread URLs	910,108
Extracted posts	8,863,628
Forums	33
Forum sections	6
Parquet shards	2
Total compressed shard size	828.2 MiB
Row groups	41
Max uncompressed row group size	139.7 MiB
Crawl timestamp range	2026-05-01T03:11:46Z to 2026-05-11T09:27:29Z
Source forum page range	1 to 15,072
Mean posts per thread row	9.7
Median posts per thread row	5
Threads with no extracted posts	1,943 (0.2%)
Rows with crawl errors	1
Longest extracted thread	53,387 posts
Mean post body length	212 characters
Median post body length	102 characters
Blank post bodies	2,303 (0.03%)
Total thread-listing views	878,540,813
Median thread-listing views	388
Extracted post good votes	10,953,215
Extracted post no-good votes	4,713,693

The largest forums by thread rows are Off Topic (526,675), Economics Discussion (109,045), General Economics Job Market Discussion (56,541), Finance Job Rumors (30,847), Political Economy & Economic Policy (22,310), Technology (20,907), Research / Journals (19,728), Econ Lounge (18,963), From the blogs (17,916), and Latest Research Discussion (14,030).

The most frequent author_role values are Economist (8,772,659 posts), Marginal Revolution (12,879), Research (11,534), EJMRBear (5,430), and Karl (5,350). author_handle values should not be treated as verified or stable user identities.

---

Dataset Creation

Curation Rationale

Econ Job Rumors is a large, long-running anonymous forum connected to economics academia, finance, public policy, and academic labor-market discussion. A structured archive supports research into anonymous professional communities, rumor circulation, hostile and informal online discourse, and the public web record of economics job-market discussion.

Source Data

Data Collection and Processing

• Tool: Custom Scrapy crawler using public Econ Job Rumors web pages
• Scope: Public forum and topic pages visible without authentication
• The crawler discovered forums from the site index, followed forum listing pagination, and crawled topic pages for discovered threads
• Each dataset row is keyed by (forum_slug, thread_slug); no duplicate keys were found in the shards
• A small number of canonical topic URLs appear under more than one forum slug, so canonical URL count is slightly lower than row count
• Post bodies were extracted from HTML into plain text with paragraph, blockquote, and line-break normalization
• Score, vote, view, post-count, freshness, and age fields are snapshot values from crawl time
• Relative timestamps were preserved as shown by the source site; absolute post creation times were not available in the crawled pages
• No systematic content filtering, PII detection, or redaction was applied to the crawled content

Who are the source data producers?

Econ Job Rumors users posting under anonymous, pseudonymous, role-like, or site-displayed handles, along with posts imported or displayed from linked blog/forum sources where present.

---

Bias, Risks, and Limitations

Content Risks

• The dataset contains anonymous rumor-forum content and should be treated as noisy, adversarial, and unverified
• Posts may include harassment, insults, profanity, slurs, sexual content, racism, sexism, xenophobia, antisemitism, political extremism, and other offensive or harmful material
• Discussions may refer to real people, universities, departments, journals, employers, hiring committees, students, and job candidates
• Claims about individuals, institutions, and events may be false, defamatory, outdated, or intentionally misleading
• The source forum has strong community-specific norms and selection effects and is not representative of economists, academics, or job-market participants as a whole

Structural Limitations

• The crawl was conducted from 2026-05-01 to 2026-05-11 and includes no content discovered beyond that snapshot
• Vote counts, view counts, freshness text, thread age text, latest-post metadata, and thread availability may have changed since crawl time
• The dataset stores relative time strings, not absolute creation timestamps for individual posts
• The dataset is limited to content discoverable through public web pages without privileged access
• Some rows have no extracted posts, and one row records a crawl error
• author_handle values are display strings from the page and should not be interpreted as reliable user identifiers
• Text extraction normalizes HTML to plain text and may lose formatting, links, images, tables, or other page structure

Personal and Sensitive Information

• No systematic PII detection or redaction was performed
• The dataset may contain names, affiliations, employment claims, personal allegations, and other potentially sensitive information about real people
• Many posts are anonymous or pseudonymous, but the content can still discuss identifiable people or organizations
• Users of this dataset should consider ethical review, minimization, access controls, and downstream harm before analysis or redistribution

Licence and Redistribution

No licence was identified for the source forum content. This dataset is therefore marked as license: unknown.