Prove it to me. If any politician or public official makes a claim, particularly one of success, I want to see the data. I never want to have to take anyone's word for it when I could see the numbers myself.

Here we can talk about ENIAC in 1952, Phil Meyer and Detroit riots and people using tape backups in the 1980s to do data work.

How do you start?

Well, like any story. I don't want to wirite about a topic. I want to answer a question. Start with a question and then find the data that will help you answer it.


How much?
How many?
How often?
Where?

How does it compare to elsewhere?
Rarely why

Be prepared that your question may not be answerable. You may have to pivot, or try a different question. Be flexible.


Places to start:
Folo old stories
Tips
Dispel rumor/myth
Steal ideas from other news orgs. (IRE)
Is x doing its job?

Va backlogs
What are veterans returning from war dealing with to get care? How long do they have to wait? Is it getting better?
Depends.

Sec 8
Aside from the corruption, how well run is our local section 8 office? Is it helping better people's lives?

Has anyone else looked at this? Nonprofits or federal agencies or academics?

Who keeps this data?

Open data is amazing, but it's only one piece. Ask your sources about what data they look at or collect. Get the name and custodian.

Also, check IRE's resource center. Look at how other news orgs have covered this issue and steal their play book.

There are developers and data geeks toiling in obscurity who are desperate to talk about the ins and outs of the systems they build and maintain. Find them!

Early. Often. Always.


FOIA the FOIA logs. Train your custodian of records.

Liens story.

The govt is not the only group collecting data. Many nonprofits and orgs collect data about themselves and even put it on their websites. Use it.

Get the data before you need it. Be prepared. Learn what's on your beat. Stockpile it.

Whether through FOIA or leaked docs.

You have some data. Now what?

Be skeptical as you kick the tires. Explore it carefully from top to bottom.

Beware of relying too much on your data. It should tip you to directions for inquiry, but it should be just one of many sources.

Sometimes it won't answer your questions, but it will give you the right questions to ask.

Is it worth your time?
What's your ROI?

What's the minimum story you can get from it?

What's the best? Shoot for the stars, but CYA. Editors want to see something for this investment of time.

If a story is too big, narrow the focus. Pick off one piece at a time.

Take your time estimate and double it.

J. LaFleur throws it out and starts over at the end.

Like a scientist, can you replicate your work and findings?

Do you have a step by step log of what you did to the data?

Don't give up. Come back to it.

York clerk had been talking about it.

Ydr had to fight for the records.

They got a PDF of thousands of pages. They kept asking.

One number in the nut.
Don't dump your whole notebook in. Use fair, telling numbers.

Don't overload.

Nerd box.
Explainer sidebar.
Blog.
Release the data itself.
Share.