WHAT CONGRESS TALKS ABOUT
WHEN THEY TALK ABOUT
POVERTY
I live in the South Bronx. Every day I go through the upper east side of New York city. Every day I see the stark difference between one of the poorest areas in America and one of the wealthiest. However, I feel like poverty is still an issue that a lot of politicians don't really want to deal with - or if they do conversations are not bipartisan and don't deal with complexity of the issue.
The Backstory
The audience for this visualization would be socially engaged, politically minded people who are interested in the causes of poverty in America and want to find solutions.
The Data
I got the data by using Temboo's Sunlight Foundations choreo [ specifically their full text search as well as their TopPhraseSources]. The data I pulled started with this new Congress, so January 2013. It took awhile to combine various data pulls so I could have together all the mentions, the Representatives district and count. I decided to do a comparison between parties so got data for both.
Once I had everything I started categorizing each statement. Some statements fit one category, some fit several. I openly admit that there can be flaws with some of the counts and categorizing. After giving each statements it's category I had to count it then pull the language for each category.
I originally wanted to a map to show the income levels in each district and then layer on top of that the number of times a Representative made a statement and which category it fit in. Geospatial mapping is really hard and it never worked out. I would like to keep on trying this though. There are still so many Representatives staying quiet.
I was also really intersted all along on the language used to talk about poverty. The ideologies can vary widely or be surprisingly similar sometimes. Looking at the Sunlight Foundations tumblr gave me some helpful inspirations on how to visual the language. I also did an initial wordcloud for each party for the total statements - but that was too broad and so I decided to do it by category instead. When doing a word cloud it takes a lot of cleaning up of language and trying to decide which words are superfluous and which are necessary. Again I admit there are probable flaws in my word clouds.
I used several tools to visualize it in the end - infogr.am, tagxedo, and many eyes. In order to give each graphic the proper design I wanted I also just did a lot of the designing myself.
The last visual about CBC activity came about after I had already gone through all the data and was doing the first 2 sets of visualizations. I knew I had to somehow show the stats I found on the CBC activity. I like the radial bar chart the best.