Due Oct 31: Rough Drafts

A rough draft does not have to have the polish of a final project, but it should be close. You should have created the visualizations that you plan to use. Your classmates should be able to evaluate a rough draft on its merits, without a guided tour of forthcoming features.

A complete rough draft includes:
+ Clean data in spreadsheets, already normalized, sorted, manipulated
+ Visualizations of the data with labelled axes
+ Captions
+ Credits
+ A headline
+ At least three links to other reporting that puts your story in a broader context.
+ Introductory text that includes information gleaned from at least one human source.

Note: you’re not required to quote your source, but you do need to be able to tell the class what insights your human source provided.

Last class?


Our last class will meet December 19, a week after CUNY students graduate. We’ll wrap up our assignments Dec 12, so we want to hear from you: what should we do with our last class.

We can bring in a speaker (is there someone you’d like to hear from?) or dig in to an advanced tool like R, d3, High Charts or TileMill.

Use the comments here to brainstorm and we’ll vote on Nov 14.

Oct 24

John Keefe, who leads the data news team at WNYC, will come talk about some of his data projects. Great posts to dive in with include:

+ Mapping campaign donations against districts
+ More campaign donations
+ Coastal Evacuation Map
+ AP Election Data
+ Live Election Data


Revise your pitch and add a wireframe/storyboard that reflects the planned layout. We use wireframe and storyboards synonymously here. We’re looking for a simple sketch (on paper, in Word, or PowerPoint, or any number of online storyboarding tools) that tells us how you intend to integrate your visualizations, words, and navigation elements.

A storyboard organizes your content conceptually and spatially. Use simple boxes to represent where your different elements are positioned in a design, and how the user navigates through the content. Here are Mark Luckie’s thoughts on sketching/storyboarding, with examples, from 10,000 Words a few year’s back.


Your assignment is to re-design one of three existing data visualizations listed below. Identify the visualization and/or design weaknesses and explain your improvements.

We’re not interested in a perfect remapping of the data, but just your overall approach of how your data viz would appear with the same set of data. Your sketch will be just an approximate of the existing data.

Deliverable: sketch on paper, preferably scanned to be submitted on WordPress and for potential presentation, but we understand if you do not have easy access to a scanner, so we will accept the hard copy.

The Atlantic: Classified Documents
Good: Teacher Salaries
National Geographic: High School Foreign Exchange

(Thanks to Jeff Heer of Stanford for design examples)


If you’re just getting started with interactive news, you probably want to start simple. You can do a lot with Google Maps, and Many Eyes is great for state and country level maps. John Keefe blogs relentlessly about his Google mapping projects, how they work, and what he’s learned. CartoDB is new and more powerful and flexible than Google Maps. It is also more complex. MapBox is even more powerful, flexible, and complex.

Great places to start thinking about maps:
+ John Keefe, Albert Sun and Jeff Larson talk about making maps.
+ Steve Romalewski lays out his critique of WNYC’s stop and frisk maps.
+ Take Care of your Chloropleth Maps
When Maps Shouldn’t be Maps
+ Amanda’s maps and mapping tags
Continue reading

Spreadsheet Skills Assignment

This week’s homework is on Lore. There are two parts, one on spreadsheets and one that asks you to look for some data that includes location information. Because we didn’t get as far into spreadsheet functions as we’d planned to Wednesday, here are some tips to get you through this week’s assignment. You can use LibreOffice Calc, MS Excel or Google Docs to work through the assignment which is harder than it looks at a glance. You’ll definitely want one of these handy:
+ Calc Function Reference
+ MS Excel Function Reference
+ Google Spreadsheets Function Reference
One subtle but important distinction is that in Google Spreadsheets, you can use syntax like B:B to refer to all of column B. In Calc and Excel, you’ll have to specify the row numbers of the cells you want to include. If the data you are interested in spans 200 rows, you’ll use B1:B200.

The small chunk of data we’re working with includes some annual rates and some hourly rates, so you need to think creatively if you want to find the average annual salary in the set. Start by thinking about what you’re trying to accomplish. Consider describing it to a houseplant out loud if you’re having trouble figuring out where to start. Continue reading