Visualizing the Influence of Money in Politics

"By making it interactive, it’s an entry point into the data, so people can create for themselves an individual and customized experience."

MAPLight.org is a nonprofit, nonpartisan research organization dedicated to providing citizens and journalists with information about the influence of money on politics in the U.S. Congress, California legislature, and soon in additional states. In a 2010 interview, MAPLight Co-founder and Executive Director Daniel Newman explained how visualizations enable MAPLight's audience to "both see the overall findings and get the deeper picture."

Tableau: What kind of data are you dealing with at MAPLight?
Dan: MAPLight.org is a big database of money given to politicians and how politicians vote. We show patterns of influence you couldn’t see before. We take all the money given to members of Congress and put it in one big database with how every member of Congress voted on every bill, so that you can see patterns of money and influence that you could never see before.

Data is a key part in educating people on what’s really going on in the country. There’s a river of money underlying everything that happens in politics, everything that happens in Congress, but a member of Congress is never going to stand up and say, “Sure, this money influenced my vote.” And an interest group is never going to tell you, “This money influenced that vote.” But the data can tell you when money influences votes. We use data to tell people in a compelling way what’s happening.

Our philosophy is we put the facts out there, and you decide. One of the amazing things about the web is that we can actually put out all of our data. We can put out summaries of it. We can put out details of it. And people can use Tableau Public to interact with it. They can both see the overall findings and get the deeper picture.

   
Tableau: What types of people follow MAPLight?
Dan: People come to MAPLight when they want to see who’s behind the news, who’s influencing the news, and who’s giving money to their member of Congress. We show them how their member of Congress votes and where they get their money from. So, people like journalists, bloggers, interested citizens, people motivated around a certain issue, bill, or person in Congress.
   
Tableau: Can you give us an example of how Tableau visualizations support MAPLight’s mission?
Dan: We were looking at where members of the California State Legislature get their money, and we found out that about 80% of the money these legislators receive comes from outside of their districts—outside of where their voters live. That’s a pretty shocking statistic, but we wanted to show people graphically what that means. We used Tableau Public to build an interactive map of the entire U.S. It has bubbles on it, showing for each zip code how big the amount of money is that comes from that zip code. You can see a big red bubble over Sacramento, the capital of California, where a lot of interest groups have their headquarters, for example. And you can even drill down to a different part of the state or a different part of the country, mouse over the bubble, and see the percentages, the amounts, all sorts of stuff like that.
   
Tableau: Without Tableau, how would you publish a finding like that?
Dan: We have published reports like that as a data table, so people can look at it on our site. Maybe they can download the spreadsheet file and sort it. We make it available to people in a tabular form and also a downloadable form. And that is of certain use, but it lacks the picture element that often draws people in.
   
Tableau: Why take the time to make your visualizations interactive?
Dan: On the web, you don’t have to publish just a summary of the information; you can publish all of the information—all of the data. So if you’re reading an article in the newspaper, of course a map or a picture is going to draw in people’s attention and give information that goes beyond just the text in the article. On the web, you can publish something static, but you’re missing this whole other dimension, which is this drill-down dimension.

Everyone coming to our website has something different in mind. People want to know the averages, sure. But, they also want to know, “What about my law-maker?”, “What about the oil industry?”, or “What about this particular bill or company I care about?” By making it interactive, it’s an entry point into the data, so people can create for themselves an individual and customized experience.

   
Tableau: Have you have any feedback from readers or from colleagues or other organizations about this?
Dan: Our findings about how much money influences politics are often shocking to people. They want to learn more. They want a way to engage. A map is certainly one easy way to do that.

Another project we did with Tableau Public was a map that looked at where California lawmakers get their money—like from business groups or laborers, and so on. One of our findings was that 40% of the money in California politics comes directly from business groups like trade associations and businesses actually writing the checks. Now that 40% number is shocking, but it doesn’t tell the whole story. There’s a lot of variation. There was one lawmaker who got 87% of his money, almost all of his money, directly from business groups. And on the low end, another lawmaker got only 4% of his money from business groups. If you look at money from unions, from labor, 16% of the money overall came from labor groups. But that’s not the whole story. The top lawmaker got 41% of his money from labor groups, quite a lot. And then there was another lawmaker that got absolutely none.

We published this information in Tableau Public. You can see the summaries, but you can also see each lawmaker with nifty interactive bar graphs for each funding category, in red or blue depending on whether the lawmaker is Republican or Democrat. You can scan down the columns and see the variation. Another finding in our study was that political parties, which contribute money to political campaigns of these candidates, only fund a handful of candidates, typically in close races. So if you look down that column, you can see certain lawmakers have a bar graph for political parties and others have almost none. That’s the kind of interesting finding in the data that the visualization makes so much easier for both us and the members of the public that see it on our site.

   
Tableau: Has Tableau helped to draw more attention to your findings?
Dan: We did a story about where members of the California Legislature get their campaign contributions and made an interactive map using Tableau Public, and the San Francisco Chronicle, a major paper here in California, actually linked to that visualization from their website and from their story about our report.

I thought it was interesting that it’s not just MAPLight that finds these visualizations of value, but also a major newspaper would find that too. And that was of value for us as another way to tell the story better, another way to interest this news publication and their audience in what we do. That’s one example of the proven value.

   
Tableau: As an early adopter, how do you see public data on the web evolving?
Dan: One trend is more and more public data becoming available. And then, the trend that’s just beginning to start is how do publishers like MAPLight.org make the data useful to people? If you have a user coming to a website, they can look at the pictures. They can look at the story. But if you want them to interact with the data, it has to be easy and obvious for them to do it. That takes a lot of design work and a lot of experience over time to build up. I think that we’re in the very early stages of web producers learning how to actually put this data out there in a way that people can interact with it in a meaningful way.

If you put a bar chart in a newspaper, that’s a format that’s been around for many decades, and we know that people can understand it. But when you put that same bar chart on the web, do you make so that people mouse over and see something? Do they have to click on it? How do you describe it to them? How do you let them dig in in a way that doesn’t become overwhelmingly complex? That’s a very hard thing to design. I think that the tools are starting to be there, but the design has to catch up.

   
Tableau: It sounds like what you’re really thinking about with MAPLight is telling stories with data.
Dan: We try and tell the story of how money influences politics, and we put all of the data out there for you to put your own judgment on that. For example, you can see on our site that members of Congress who have voted with drug companies have gotten more money from drug companies. We have custom-created and custom-programmed bar graphs and timelines that make it easier to see and to dig in to this data. Where Tableau Public comes in is where it is not possible for our small programming team to create every possible visualization.

We use the Tableau Public library to build in another level of complexity—one that’s well suited for all the different findings that we have. Sometimes we’re looking at one committee of Congress. Sometimes we’re looking at one member of Congress.  Sometimes it’s one industry. There’s no one size fits all way to say, “What is going to be interesting and where is the diversity happening?” The diversity of visualizations in Tableau Public and the flexibility of it is a real asset that we can actually publish visual information about some of these stories beyond the standard visualizations that are available throughout our site.

   
Tableau: How did you find out about Tableau Public?
Dan: A member of our staff attended a class on data visualization, and many of the examples were created with Tableau Desktop tool. Then I attended a workshop in San Francisco where Tableau was exhibiting. The representatives were very responsive and actually loaded in some of our data right there on the spot and showed how it could be visualized.
   
Tableau: How easy has it been to publish these interactive visualizations?
Dan: The technical side has been very easy. We figure out what data table we need. We upload it into Tableau Public. And then it creates essentially a widget we can put it on our site. We’ve gotten a lot of help from the staff at Tableau to take a simple visualization and make it multi-part and more interactive and more compelling. That’s something that’s going to take a lot of more practice internally for us to figure out how to do, and that’s been a very necessary part of getting the high level of results that we have.
   
Tableau: Do you feel like Tableau is accessible for people who aren’t programmers?
Dan: One of the things I like about Tableau Public is our small staff here of about half a dozen—all of them can potentially use the tool. I mean, they haven’t all learned to do it yet, but it doesn’t take programming.  You load the data, kind of like a data spreadsheet or a Google spreadsheet, and you can choose from pictures. The interface can be complicated, but it can be mastered by someone who’s non-technical. That is very appealing. We’re educators here, as well as technical people, so it’s really important that people who don’t have technical skill can actually use this to get the stories out.
   
Tableau: Is there anything else you’d like to say about Tableau or MAPLight’s work with data?
Dan: One of the trends we’re seeing in the web and in society is greater transparency and that means greater transparency for data. You have the government releasing more data. You have private citizens putting together their own databases just out of interest. So that data becomes more available. But it’s not any easier to know what to do with it.

The other trend is the increasing use of the web, which of course is only going to continue to increase over time. People are becoming more sophisticated, not just using email but doing more sophisticated searching and using the web for different purposes.

So we have this intersection of more sophisticated users and more data available. Tableau Public is at that intersection. It takes these data sets, it makes them more visual, and it’s a way to interface to these more sophisticated users. It’s certainly not the only thing going on in the sphere, but the concept and execution are very important components in these two trends coming together.