Skip to content
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
31 lines (23 sloc) 2.77 KB


We welcome all those interested in contributing to this analysis of media selectivity. We’ve listed out a few concrete ways you can help with a focus on data collection, categorization, and analysis.

While many of the ways to contribute will appeal to hackers, engineers, data scientists, there are a number of tasks that don’t require such specialized skills.

Ways to help:

  1. Determine the amount of coverage the same event will get in different communities/countries; this will require looking at various sources in the days after an event happens. Examples include

    A. Terrorism
    B. Drone Strikes (especially one with civilian fatalities)
    C. Hate crimes against a specific ethnic group (a US anti-Muslim attack’s coverage/sharing in the US versus its coverage/sharing in Saudi Arabia)
    D. Natural disaster
    E. US spying of other countries vs. other countries spying on the US

  2. Assemble media coverage data for other sources: One key critique of our analysis is that we’ve chosen a single data source – while conjecturing that the bias found there is found in all forms of media. Help collect and categorize content over a time period for any other media source of your choosing including:

    A. Any publication in other regions of the world (e.g., Times of India, Le Monde)
    B. Your Facebook feed
    C. Your Twitter feed
    D. Any industry publication
    E. (For coders, you can build an automated data crawler – and for others, you can manually record)

  3. Identify and assemble a form of objective data to compare media coverage to: Our analysis uses only a few forms of objective data (national population/wealth statistics and deaths) – and media bias can be teased out further by comparing to other forms of objective data that you can assemble

  4. Look at extent of coverage of certain events; for example, see what percentage of articles were about the 2016 US Presidential election vs. all other topics

  5. Apply sentiment analysis to media coverage: Our analysis looks at article counts – and neglects a key part which is the content itself, such as the tone of coverage; for example, there exists widely used metrics to determine how partisan a media source is

  6. Develop a method to “unbias” your favorite media source of choice or measure the various “food pyramid” proportions; take a biased source, and create a method that would reduce the bias, however you define it

  7. Identify and help correct mistakes in our analysis: This analysis was created by one person working over several weeks – and required a substantial amount of judgment; point out mistakes and/or ways to improve further

  8. Want to suggest another task? Let me know

If you have specific skills, drop me a note as well.

You can’t perform that action at this time.