This piece is adapted from the Tow Center for Digital Journalism’s weekly newsletter.
Sarah Stonbely, PhD, is the research director of the Center for Cooperative Media at Montclair State University. This week, in collaboration with the Tow Center, she’s publishing a report on the state of local news provision in New Jersey. From her research, Stonbely argues that studying the local news crisis is best done by looking at local news provision, or the actual coverage areas of a local news organization. By analyzing New Jersey from the municipal level, Stonbely and her research team were able to create a more nuanced map of where news is coming from in the state, and where it may be overlooked. They also integrated municipal-level median household income, tax-spending data, and population demographics to better understand who was receiving or else missing out on local news. Stonbely spoke with the Tow Center about the study and its implications for the future of local news funding, resourcing, and coverage.
What inspired this project?
Stonbely: My work at the Center for Cooperative Media is focused on what’s happening on the ground in local media. But it’s also happening in this moment of intense change in the local journalism industry. Our research agenda was really to get a handle on what is happening in the local journalism field in New Jersey, but we didn’t have data on the number of outlets at the municipal level. And when we looked around, no one had this data. So there was a real need. Penny Abernathy has mapped the county level, and that works for states that are less densely populated.
New Jersey has acted as a sort of test kitchen or a Petri dish for a lot of the funding that’s gone into local journalism innovation. But even before the Center really became what it is now, there was money going into local journalism trying to innovate the model.
Why is it important to zoom in on the municipal level in a place that’s as populous as New Jersey? And could the zoomed in model be applied in a less dense place?
Stonbely: Life is really lived at the municipal level in a place as dense as New Jersey and elsewhere. Because of the way that the local government is structured, the decisions around taxes and public services are made at that level. Some decisions are made at the county level, but there’s so much variance across counties. For example, we’re in Essex County, which includes Montclair, which is very affluent, sort of feels very progressive in certain ways. And then there’s Newark, which is much less affluent, and has a lot of problems around policing. These are very different types of communities, but they’re both within the same county. So to generalize “Essex County” just doesn’t work.
We wanted this to be a pilot that we could apply elsewhere. I’m really curious about how it’s going to map onto a place where the county and the municipality don’t mean the same thing. A lot of states don’t even use “municipality.”
The report is data-driven, but, you also crafted a logical narrative alongside that data. Was there any reporting or on-the-ground research with these organizations?
Stonbely: No. This phase of the project was entirely data-driven. The larger project, which is multi-phased, will include interviews, especially with the news conglomerates that we’ve identified. There are dozens of local conglomerates in New Jersey, and that’s fascinating. We’re also going to talk to audience members because you can talk all you want about local news, but who knows who’s actually paying attention?
The benefits of using the data that we did is that you can look at this wider landscape and do statistical comparison, which is really difficult to do with more qualitative data. The drawback, of course, is that you lose some of the nuance or specificity. But really the goal was to have that bird’s eye view and then, in subsequent phases, to dive in and dig down with those qualitative methods.
How did you conclude that municipal spending and median household income were the most important factors when analyzing local news coverage and reach?
Stonbely: I think one of the most intuitive structural correlates for local news provision is the affluence of a community. We looked at 565 different municipalities, and sure enough, the affluence of a community was statistically correlated with the number of outlets. That has a lot of implications, especially when local journalism is looking for new business models, a lot of which are audience-revenue driven. You can’t really count on communities that don’t have excess wealth to be able to put money into news outlets, right? So the importance of promoting different policies—like giving consumers credits if they want to become subscribers, which will then be reimbursed by the government, local or otherwise—is clear.
A lot of these new business models have been premised on the idea that, okay, audiences can just step up. But in some of these communities, that’s probably not really realistic.
In terms of municipal spending, the idea is that communities [in New Jersey] are spending all of these tax dollars, but no one is reporting on it locally. No one’s attending the meetings where they’re discussing how to spend the money. In our research, we found that in 2016, New Jersey municipalities spent over $15 billion. Half of those same municipalities were only covered by zero to two outlets per capita.
Corruption is a huge problem, and no one’s reporting on it. And certainly there are exceptions of great local reporting being done, but I think the findings jumped out to me because once I sort of did the math [on tax dollars being spent], I was like, Whoa.
But an important methodological note: the numbers were weighted by population, which is certainly common among economists because it can make a huge difference. So if you think about just using raw numbers, it would be the equivalent of saying Big Sky, Montana, has 10 local news outlets and Newark has 10 local news outlets, therefore they have the same amount of news coverage. But there’s no equivalence there because of population density. There’s also more money being spent in Newark.
You concluded that the local news crisis is perhaps not as dire as we think because a lot of research doesn’t weigh coverage against population.
Stonbely: I’m way more optimistic about the local news situation than a lot of other people who write about it because of exactly that—the weighing of population. I think there were only four municipalities in all of New Jersey out of 565 that had zero outlets. So very few true deserts. Twenty-nine municipalities had zero to one. Now, there are a lot of caveats to all of this because our analysis was complicated. Using coverage area instead of raw number of outlets or geographic location of the outlet was one of the innovations of the method.
Now the second methodological note I want to make is about “local news originators.” The number that I used for all of the statistical analysis, in addition to being weighted, was also only focused on the local news originators, of which there were 565. That is the same number of municipalities, which is horribly confusing and inconvenient. But that’s just the way it is. So this does not count most of the radio stations other than the public radio stations, which have their own newsrooms.
So the raw total number of local news providers serving New Jersey is 779. But a lot of studies have found that some providers, especially in broadcast television and the radio, are not going to be the originators of the news. With that, it’s going to be the local newspaper who goes out, sends the reporters to the municipal town meetings, sends the reporters to talk to the school district, all of that local news gathering. And then the TV or the radio station will essentially read the story, and just rewrite it into a much shorter blurb, which they’ll then read on the air or on TV. So I chose to focus on news originators because I didn’t want to artificially inflate the number of outlets, because not all of them are producing the news.
What do you anticipate from the next phase of research [which is a content analysis]?
Stonbely: Without looking at any of my data, which is still being gathered and scraped, I’m anticipating that it’s going to be a much less rosy picture, actually. If you’re basing things on self-reported coverage areas, an outlet has a lot of incentives to say they cover more towns than they actually do because they’re usually giving that information to advertisers. So I really anticipate it’s going to be a much more desert-y type of landscape.
With your findings in mind, do you think we need to take a different approach to ongoing legislation regarding local journalism?
Stonbely: I hope that this research can influence policy around local journalism funding. We were able to give this research directly to the board members of the New Jersey Civic Information Consortium, which is really exciting. If we could have any impact on directing money toward municipalities, I think it really helps the roadmap toward more equitable local news coverage. If you can say, “Okay, in general, affluent communities don’t need as much funding or help or additional outlets because they’re having more audience revenue, let’s really try to direct money toward underfunded communities, or undercovered communities, or where the Hispanic population is a certain percentage.” That very specific roadmap is my ultimate hope.
Do you think there are any implications or recommendations from your research that can be directed toward journalists or editors?
Stonbely: I think that any outlets that have discontinued looking at municipal spending might think about bringing that back. In general, I think journalists are just doing the best they can. And in many cases, they’re doing a great job. But other research has found that even just the presence of municipal spending coverage acts as a check on malfeasance. So looking around at the community and seeing what’s not being covered, paying attention to municipal meetings and things like that.
Is there any element or part of this that you would do differently?
Stonbely: Oh boy. Well, from the very beginning when we were building our initial lists and database, I probably would have kept track of more variables about the outlets: ownership, whether an outlet is owned by a conglomerate or not, stuff like that. I didn’t talk about that in this paper. There’s also a lot of stuff about the makeup of the newsroom in terms of race and gender that I wish we had. I talk a lot about the importance of mapping by coverage area, but I think knowing where the outlet is actually based and comparing that might’ve been useful as well. We do have a little bit of data about the languages that news is published in, and we’re now going back and filling in some of those holes.
I’m really happy with all of the data we gathered, the structural correlates, and all that sort of stuff. But you always want more.
This conversation has been edited for length and clarity.Sara Sheridan