Data analysis N Ways

Session notes, data, instructions and examples for a hands-on workshop on using a diverse set of tools and practices for journalistic data analysis. For SRCCON 2016 http://srccon.org/sessions/#proposal-318397

Workshop agenda

Describe the task, break into groups (5 min)

Place signs around the room where people can sit because they're interested in working with a specific tool. Some examples might be:

• R
• Pandas
• Agate
• Excel
• SQL

Have a few blank signs so people can have the option of forming a new group for a tool that we haven't mentioned.

As participants trickle in before the session, tell the participants that they'll be doing data analysis for a news story. Direct them toward the part of the room with the sign for the analys tool they're interested in. They can use a tool they're familiar with, or one that they want to learn. It doesn't matter.

Facilitators will encourage groups to take a mob programming ("All the brilliant people working at the same time, in the same space, at the same computer, on the same thing") approach, being sure to switch up the "driver" every few minutes.

Analysis (25-30 min)

Point users to the analysis steps document in the repo, as well as the raw data and tell them to get going.

Retrospective (25-30 min)

Each group will be given some prompts to reflect on the work they've done. Each group will pick a representatives to answer one of the questions in front of everybody.

• What worked well with the tool you chose?
• What didn't work with the tool you chose?
• What was unclear about the analysis script?
• What additional questions would you ask of the data?
• How would you share your analysis with a colleague (technical and non-technical)?
• Does your approach allow for good documentation?
• How would your approach handle new data?
• What libraries did you use with your tool?
• How would you publish this analysis?
  • How would you make it into a graphic?
  • How would you share the data if someone else is doing this?

As people are sharing

Participants will be asked to send a pull request or email their analysis scripts/workbooks/files to include in this repo.

Workshop materials

• Post-it notes
• Paper, cards to make signs to identify groups

About the data

This data was used in the process of reporting this story and making this interactive.

It is based on the EPA's Lead and Copper Rule.