GlobalVoices - A Sample GenderTracker Client

GlobalVoices is an international community of bloggers who report on blogs and citizen media from around the world. As a first attempt to test the GenderTracker system, we are working with GlobalVoices data to evaluate its content's gender diversity.

Note: in 2018, several years after publishing the original Guardian article, we removed the data from this analysis upon request from some participants in the project. If researchers are interested in re-analyzing this project to double-check our results, please contact J. Nathan Matias. With approval from a university ethics board, this may be possible if it falls within the original purpose of the data collection.

Requirements

1. get jruby 1.6.3 by your favorite method of choice. I recommend rvm.
2. run gem install bundler
3. run bundle install
4. Intall redis
5. Run redis.
6. Make sure the genderTracker server is running.

Folders and their meaning

There are several key folders of interest:

1. The data folder contains the available data for global voices. Note there are two collections in place:

• sample - a small sample of the articles (about 1800)
• full - 9 years worth of content

This folder contains the raw data. In this case, an xml feed.

2. Once articles get processed, they go into the articles folder under the apporpriate collection name. Note we don't check those files in since the full collection especially is rather large. You can generate them yourself.

Available Scripts

prepare.rb

This script converts the xml data into the article.json files. It doens't talk to the GenderTracker service yet. To execute it, run:

bundle exec prepare.rb sample (or other collection name)

process.rb

This script actually sends the availble article json files inside a specific collection to be processed by the main genderTracker service (so make sure you have that running.)

The script terminates when all articles have been processed.

Note it requires redis to run, so make sure both the genderTracker service and the process.rb script are pointing to the same redis queue. You can change that in the config.yaml.

Note that the genderTracker service actually modifies the filepaths that it recieves, so it must have read and write access to them.

To execute the script, run:

bundle exec process.rb sample (or other collection name)

aggregate.rb

This script is responsible for assembling some csvs about the processed files. By deault it creates all its files in the assets folder. Two files are generated:

• *_all.csv - a csv of all the articles, their pub date and the results of the metrics.
• *_names.csv - A unique list of authors that contributed to the above set of articles and our guess as to their gender.

To execute the script, run:

bundle exec ruby aggregate.rb sample (or other collection name)

bundle exec ruby aggregate.rb sample CategoryName (any category that appears in the category property of articles.)

Licensing

The project is duel licensed under the GPLv3 license as well as the MIT license. Pick your favorite.