Resources

About our project

In addition to reading about our initiative on this website, here is a one-page NewsQ Brief that described the state of the project in April 2020.

Principles and Recommendations

Panels. In its effort to support broad constructive engagement on the question of news quality, the NewsQ Initiative hosted a series of panels. NewsQ Panels were composed of journalists, journalism experts, and technologists, and review specific cases of news ranking and recommendation to understand whether generalizable guidance or principles regarding quality journalism—interpretable both by algorithms and by journalists—might emerge.

Data collections

NewsQ has collected two types of information: potential “signals” of quality related to journalism and examples of feed results from platform news products. We worked with our collaboration partner Knowledge Futures Group who maintains much of this data.

API. As of 2020, the NewsQ Database contains over 100 units of information on over 12,000 news and information sites. Some of these units of information may be potential “signals” of quality journalism. This database, which is mostly focused on English-language US readership, was open for testing purposes to a limited number of invited researchers and participants and closed down in 2022. An archive of the data has been transferred to Knowledge Futures Group.

Some documentation regarding the former API can be found below:

Feed results. Our project collected specific results from news feeds across major platform news products by Google, Facebook, Twitter, Apple, and Microsoft for an initial qualitative review during our 2020 panels. The dataset, gathered across dates from February to June 2020, provides an illustrative rather than systematically sampled view of possible results. In addition to screenshots of news feed results, the results are accessible in an Airtable with additional contextual information about each of the top five articles, including the article URL.

In addition, in 2021, a dataset of results from Google News, Microsoft News, and Twitter Explore was also collected for analysis.

Prototypes

News ranking prototypes. Code and example datasets from winning contents of the NewsQ NewsRank for Social Good Hackathon can be found at Devpost.

News within online information networks. This gap analysis on vaccine safety information on English Wikipedia began in 2020 and will be further developed in 2021, as well as analyzed for its handling of news sources.

Also see:

Reference Materials

  • Orientation Guide on Algorithms and News Ranking: an orientation to help newcomers to algorithms understand the challenge of news ranking and recommendation systems better. 
  • Orientation Guide to Platform News Products: Provides overviews for eight news products, mostly from the perspective of the goals expressed by platform owners themselves, with background and process for each product (vis-a-vis news).
  • Reference Guide: News Genre: Includes examples of classification systems, including Opinion.
  • “Journalistic Values and Expertise in Platform News Distribution: The possibilities and limitations of participatory panels for algorithmic governance” (DOI) (Accepted Manuscript): This paper in Journalism Studies by Connie Moon Sehat describes the methodology of the NewsQ panels project, as well as outlines future challenges.
  • News Quality in the Platform Era: Research writings from a NewsQ-supported Social Science Research Council workshop are forthcoming in 2022 by Routledge.

Reading Library

Our Zotero-powered library of relevant reading materials (no longer maintained) if you are interested in browsing.