Feeds:
Posts
Comments

Posts Tagged ‘data visualization’

I taught a class Monday in data visualization for Josh Grimm’s In-Depth Reporting class at LSU.

I’m no expert in data visualization, but I studied the use of interactive databases for the American Press Institute in 2008 and my students experimented with a variety of data viz tools last spring in my course on learning interactive storytelling tools. (I’ll add some links to the students’ posts on data-viz tools later, but I want to get this published now and I won’t have time to add links until later.)

My point in this class is that you can tell stories lots of different ways using data, and that you can teach yourself pretty easily how to use some effective data viz tools. I admire the skills of some data specialists I know, and hope some of our students will follow them into that specialty. But I hope every student (and professional) journalist develops data skills to find and tell stories routinely.

Examples I used in the class (and a few I didn’t have time to use):

Thanks to Kyle Whitfield, Mark Lorando, Tom Meagher, Maryjo Webster, Daniel Tedford, Kevin Dupuy and Michelle Rogers for providing these examples.

I collected information from the students using a Google Form and used it to create some data visualizations about the class using Infogr.am and Google Maps. I was running out of time and rushed through these pretty quickly, but you can make pretty simple graphics quickly using these tools. I elaborate a bit more here on some of them.

I wasn’t able to embed the resulting Infogr.am graphics in my free WordPress blog (they should embed on most websites). Here are some screen grabs of the graphics (with links below to the interactive versions):

Infogram devices

You can see the interactive version of the graphic on devices here.

This pie chart, I noted, would be more effective with graduated shades (perhaps yellow to red) than the random colors assigned to each number:

number of devices

In a graphic about the students’ use of social media, I tried different data viz tools offered by Infogr.am. This line chart didn’t work for me (though it might work for other detail). An effective graphic makes a point quickly and this one requires some study:

infogram line graph

This horizontal bar graph also took a bit of work to understand, but quickly shows that the most popular social tools with the students are Snapchat, Instagram, Facebook and Twitter, and that the students aren’t using Foursquare at all. The graphic on devices was filled out later, when I had 26 responses instead of 24.

I deliberately didn’t update this because it actually illustrates some points you need to check in doing data visualization: The data need to be accurate. My first take of this didn’t have fully accurate data: You can see that I only have 23 responses, instead of 24, on Snapchat and Instagram. Actually, I had 24 responses at the time, but failed to double-check my data before uploading it for the graphic. These are the kinds of errors you need to avoid and double-checking you need to do both before uploading data and after finishing a visualization project.

infogram bar chart

The most effective graphic on social networks, I thought, was this layered pie chart, where you can (in the interactive version, not the screengrab below) see how differently students use the social tools. It would have been more effective, though, with a gradual color scale, perhaps with yellow for 1, orange for 3 and red for 5, with shades in between at 2 and 4. But I was trying to show how quickly you can make a simple graphic. That’s the first step in data visualization. I’d expect such improvements in subsequent projects.

infogram pie chart

Moving to Google Maps, I quickly imported information from the spreadsheet of student responses to create a map showing where the students were from (that embed works here):

During the class, Deanna Narveson did a quick data viz project on social media engagement by Louisiana gubernatorial candidates:
https://public.tableau.com/javascripts/api/viz_v1.js

Dashboard 1

Here are my slides from the class:

Advertisements

Read Full Post »

A complex story should not be challenge to the reader or viewer, however challenging it is for the writer. Careful work in organization of your reporting, digital production and writing will help readers make sense of stories that deal with cumbersome economic or technical issues, or with soap-opera tales that present multiple characters and confusing turns. These techniques will help keep the complex story clear.

Use digital storytelling tools

Reporters with long print experience tend to think they need to squeeze everything into the text story that they love to write. Digital First journalists need to think about the best tools for telling each part of the story.

The bigger the story, the more different digital storytelling tools you should consider. But an important part of organizing the story is to avoid overwhelming the reader or viewer with every fact and every tool you might use. Choose the most important information and then decide which tools share that information the best. Much of the success in a complex story is in those difficult decisions of what to leave out.

Videos and photos

For the strongly visual aspects of the story, use the best visual storytelling tools. Instead of writing a sidebar on a topic with visual appeal, or squeezing it into your text story, make it a Tout video or a longer video and give it prominent play in the package.

Or tell a story in a photo gallery. If a picture is worth a thousand words, an effective photo gallery saves you a lot of writing and lets the writer concentrate on the points that are best conveyed in words. (more…)

Read Full Post »

I’m a great fan of data visualization, and it’s not something I have done much of myself (I can make an Excel graph). So I will be interested to follow Visual.ly, a fascinating new project that showcases data visualization efforts (check out Golden Parachutes and the History of Location Technology) and promises to develop and offer some dataviz tools.

For now, I used its Twitterize tool to visualize my own Twitter use:

(more…)

Read Full Post »