Mike Skoler
Managing Director of News
Minnesota Public Radio
"Budget Balancer"

In the interest of full disclosure, I should mention that The Seattle Times earlier work in interactive journalism was one of the inspirations for what Minnesota Public Radio is doing.

I came to Minnesota Public Radio seven months ago with a mandate to invent a new way of doing journalism that brings the audience into the news process. We call it interactive journalism, but, when we use that term, we mean it a little differently than simply presenting information in a way that directly involves the audience and requires audience participation.

At Minnesota Public Radio. we create simulations, we do games, we do surveys, we use online tools to engage and converse with our public, with our audience, but those are tools, even if important tools. At Minnesota Public Radio, interactive journalism is something different for us. It's about gleaning knowledge and insight from the audience. It's about improving our daily journalism and the way that we actually set the news agenda in an attempt to be more of a leader of news coverage in our region.

We start with the assumption that, on any given story, there is someone in the audience who knows more about it than we do. That might be a business leader who won't go on the record; it might be a middle manager who would be hard to identify in the organization; for a story on adolescent mental health, it might be the parent of a teenager who's mentally ill. Our goal at Minnesota Public Radio is to tap into that audience expertise and do it in a way that is large, engaging thousands of people, and directly impacts the journalism that we do, the quality of the journalism and the way we make story selections.

It sounds noble, A, but if you're the skeptical sort, and I'm assuming you are, since you're journalists and educators, you're thinking, "How do you do this without swamping a newsroom with information from tens of thousands of folks? And how do you do it without loosening the standards of journalism that we all strive to maintain?"

I'll admit off the bat, we're not fully sure how to do that, and that 's what we're trying to develop and certainly doing in our series of experiments. What I'd like to do is talk about some of those experiments, but mainly describe our most ambitious one to date, which is the MPR Budget Balancer.

Minnesota, like many states, had a huge looming deficit, $4.2 billion, and it was a major regional news story throughout the spring. Budget debates are tough to cover, I'd assume some of you have had experience with that, because budgets are so big and complex, that what you tend to do with a budget is break up it into pieces, and you talk about how spending cuts or extra spending might affect health care, might affect highways or whatever. It's really hard to present the budget as one whole piece and really help the public feel the tough trade-offs that come with trying to set a budget, especially in the current environment where it's all about reducing deficits.

So we wanted to know both how people balance budgets but also how they think through the budget trade-offs. Our aim was to engage lots of people in this and, by figuring out how they think through the budget, actually pick stories that would be most meaningful, most important to our community. So we created a simulation, and called it the MPR Budget Balancer. I won't show you all 19 screens of the balancer, which provided people with 62 different choices on how to balance the budget, both cutting spending and adding revenue. It takes about 10-15 minutes to get through it, and we were really concerned whether or not people would stick with us that long.

This first screen was kind of a setting of expectations and a bit of a warning that it would take a little bit of time and that they didn't have complete freedom in this: There would be a specific set of options that we provided, and they would have the opportunity to present others later.

We required an e-mail and a zip code. We thought about not requiring that to try to get the most people in, but we really wanted this to be a bit of a transaction where, in order to play with us, we wanted to find a group of people that would engage with us regularly in our news coverage. We asked for some demographic information, and we were pleased to find that 98 percent of the respondents -- this part was voluntary, the demographic information -- 98 percent gave us some information and more than half gave all of the information we asked for. We asked for simple things: We asked for age; we asked for where you lived, whether it was city or suburbs or out in a rural area and the like; what gender you were.

After giving information about the current budget, we offered users some big swipes at the deficit. For instance, Option 2, which you probably can't read, you could start with a common political move and use some creative accounting to delay payments on state expenses. This proved very popular (laughter) as you can imagine, and popular with the politicians, saving $550 million.

Then we offered specific budget cutting options by category for more than seven screens. Here's health and human services, and each page, as you'll see on the right, starts with headlines, which provide basic facts about the category and it might explain where Minnesota compares with other states in terms of spending in that category. If you choose an option, like the first one, to cut basic health grants for low income adults and families, two things happen. First, you'll see the deficit clock, which is a little hard to read here, changes, it now says $3.6 billion instead of $4.2 billion, so you're cutting the deficit, your goal is to get it to zero.

Secondly you get this big red box that says "Look Out" and talks about the possible consequences of the choice that you've made. In this case, up to 33,000 Minnesotans could lose access to state-subsidized health insurance. That's the standard form for all of these options.

From cutting, through these seven screens, we move to options for adding revenue, and the pie chart explains the breakdown of where money comes from now and at any point you could click on a category, like sales tax, and see an explanation, in this case saying that the current sales tax is 6.5 percent for some goods but there is no tax, for example, on clothing. Then we offer two pages of revenue-raising options, so we might click on the first option to raise income taxes and & we find out that Minnesota might move from the eighth highest taxing state into the top five.

When you balance the budget, you get a pop-up box that provides you with congratulations and gives you the option to continue cutting, if you want to actually create a reserve, or review your material, or to submit. At any point, on every page, you've got a submit button at the bottom, where you can get out of the simulation and go to the payoff, which was an immediate comparison with the governor's proposed budget, line by line. The checks show where you agree with the governor, and where there's no check you have a difference with the governor's proposal. The governor's proposal was really the working document for our budget debate.

We provided an easy-to-print version of this, which was very popular with secondary school teachers, and we'll talk more about that in a second. Then we provided a visual look of how your plan affects the overall budget pie compared, again, with the governor's plan. In some other screens, the balancer shows you how the state ranks on various quality-of-life measures, from educational achievement to level of taxation, and then asks people to consider how their choices on this budget might, over time, affect the state's future and its ranking on those. I won't show you that, for time.

Finally, people could tell us if they were surprised by their choices. Seventeen percent said they were. Many of them said they were surprised that they had voted not to raise taxes but found themselves raising taxes because of the trade-offs in this.

We provided the opportunity to comment on the balancer and its options, and one in every five people provided comments about the simulation. We recorded every choice every person made for analysis and that's what made this different than just an interesting way for people to engage and learn about it. We wanted to learn from what people were doing.

The project took a lot of folks. We did it in 3 1/2 weeks, and it was tight, and things were changing on us. The governor's proposal when we started wasn't even on the table yet, the numbers changed through it. We ended up hiring a software firm in California with a lot of experience with educational software to actually do the total of this, because our new media folks just didn't have time to do it and it was quite intensive, but we consider it highly successful.

Numbers are higher than what Eric Pryne was talking about. We had the advantage that we had radio to actually promote this and we promoted it heavily. After every budget story we did, a host would announce, "Go to the MPR Budget Balancer, you take a crack at the budget," so we were able to engage a lot of people; 19,000 checked out the balancer, which for us was a huge response.

People wrote comments thanking us, many organizations promoted the balancer or linked to it on their sites, such as the Minnesota League of Women Voters. A county commissioner in a public hearing actually stopped and urged all of his colleagues there to go on the MPR Budget Balancer; you couldn't ask for better publicity. "Come create your own budget. You come to us with your plan."

We ended up with 7,000 people who actually stuck with us for the 10-15 minutes or longer, and the 7,000 people actually generated 11,000 plans. Many of them actually submitted more than one plan. We got 7,000 by removing the doubles and by removing people outside of the state and the like. One expected, but still very nice surprise, and Eric referred to this, is that we attracted a somewhat different audience. On the right you have the typical public radio listener and on the left the Balancer users. Generally just 10 percent of public radio listeners are 30 or younger, 43 percent of the balancer users were 30 or younger. Obviously the medium has something to do with that -- computer use.

Also we reached a lot of school-age kids. 0-18. It says 17 percent, I think the final tally was actually 20 percent. We promoted this heavily to secondary school teachers; many of them assigned it for extra credit and assigned it in class. And [we reached] a large number of 19- to 30-year-olds. We feel this fits with the notion that there's a new generation of folks who are used to being engaged with nearly every aspect of their lives, whether it's planning their finances and trading constantly or whether it's actually interacting with news and public information and being a part of the news that they receive.

And the Balancer strengthened our journalism. From user comments and the popular choices that we saw, we got an outside perspective on our coverage and a sense of some potentially important issues that weren't being covered. A popular choice for cuts on the Balancer was a health care rationing system, similar to what Oregon had done, which, instead of cutting the rolls of who could receive health insurance, it cut some of the less cost-effective services; it reduced services. So we did a little story explaining that before the issue was actually proposed in the legislature.

Taxes were another big issue: 85 percent of the people who participated actually ended up using the revenue-raising opportunity. As Eric mentioned, this is a survey not a poll, so we went out and commissioned a poll to see if what we were hearing, what we were seeing in the Balancer, was actually an undercurrent that wasn't being recorded. What we did in the poll was we provided some of the trade-offs, and we found that when we asked people in Minnesota, "Would you cut taxes or would you raise revenue?" in general the majority said they don't want taxes raised, let's cut spending. But when we actually presented specific options, we found that the majority of people actually would raise taxes in order to preserve some services, up to 70 or 80 percent, and we actually found that 70 percent favored two specific ways to raise taxes, so we were able to actually write that story thanks to the Balancer. A number of stories came from it.

This wasn't the end of the story. Through this project, we ended up with nearly 5,000 people who said they would engage regularly with us in helping us with news coverage. During the Balancer process we did a series on eating and overeating, on obesity, and we got about a 2 percent response rate from folks, which seems small to us, but was helpful, People shared some very personal stories, and I'm not going to play the CD I was going to play for that because I'm running out of time.

Since then, we've been doing surveys of this group of 5,000, and we asked them a follow-up question at the end of the budget debate. This time 18 percent of them, 860 of the 5,000, actually came back and gave us detailed reactions. This is an example of the report we get, not the survey, and we blacked out the names at the top, those aren't bars of percentages. Each question we ask would get a response and we would put those into reports that we would then provide to editors and reporters doing stories. In some cases, we went back to specific people who had information that identified that they had expertise; we used them in stories.

We also currently have a survey that's being used nationally, as part of a collaboration on public radio stations and web links for special coverage appearing this November. The series is called (inaudible), the coverage is called "Whose Democracy is it?" We have this survey up, and this actually resides on station web sites, and now National Public Radio has it, and when people fill out this survey, and it asks them questions like, "Do you participate in sustaining democracy," "Give us a story of when it's worked for you or hasn't worked for you," in fact it goes over a national database that we're using to help guide the editorial material for the special coverage, it goes back to the individual station the user kind of came through. So local stations are getting a sense of what's important in their local area, we're getting a sense nationally of what stories and source ideas are nationally. We're using this very regularly, this interactive survey tool that we've created. We've kind of taken where we've gone from the budget balancer, we've taken that group of people and we're now in regular contact with them, on this and other things. We plan some other group projects, but with these its constant interaction we've got.

Question:
(inaudible)

Answer:
We didn't seem to, but we were very clear from the start that we didn't want to publish any of these results, and we never spoke of percentages of people who chose anything because we just wanted to make it clear that this was to help inform our judgment and our insight and our editorial sixth sense, it wasn't to actually report on this and that's why a key part of what we do is, if we have a hypothesis based on this we create a poll to actually get some scientific results. So we didn't sense that and there was no agenda, there was no reason to kind of come in and skew this because it didn't appear on the air. Yes?

Question:
(inaudible)

Answer:
What we're doing is, we're promoting it on air, we're promoting it on our web site and with some of these e-mails to these folks to alert them that there's a new survey up. What we're trying to do is actually keep building that base. It's not just about technology, we actually have a salon program where we're going to start going to peoples homes to have discussions around interactive journalism & Minnesota state fair is coming at the end of August, we're going to have a booth there where people can come in and talk about a particular question for a minute or a minute and a half to get out the idea that we're trying to listen, we're trying to gather people, as well as people not connected to a computer with an 800 number and the like, so we try to keep building that, but this base that we've got is a base that we're getting information now about occupation and the like so that we can actually send out selective e-mails.

Question:
There's a lot of content on this site, how did you create the content, how did you verify...(inaudible)

Answer:
A lot of hard work, and that's why there's that big team of people up there. What we ended up doing is, besides sending two reporters out full time to gather this and analyzing all the documents from the governor and documents from other organizations in the state, once we had an initial draft of the Balancer we started testing that among experts and people of different political parties to try to catch where language seemed to be biased or facts were in dispute and then, as good reporters, we would go out and actually try to determine our best sense of what the true facts were about impact or  there were even disagreements about the governor's proposal and what it actually did that we had to sort out, it's just a lot of basic journalism. It's scary, because you're putting out a lot and the ability of people to attack you is great, so there's a concern. Thank you. (applause)

Back to AEJMC Interactive Journalism Summit 2003