How Giant GateHouse Media Performed in Harsh Duke Study of Local News Sites

Share this:

The desertification of local news is creeping its way across America—not only in poorer, rural areas, but also in suburbs and even cities. That’s the theme of much recent academic research, anyway.

It is the strongly implied message of the massive new study “Understanding Local Journalism: News Deserts, Journalism Divides, and the Determinants of the Robustness of Local News,” produced by a team of researchers at the DeWitt Wallace Center for Media & Democracy at Duke University and discussed here recently. After looking at how news sites of newspapers, radio, and TV stations and digital “pure-plays” performed in a hundred randomly selected communities, the study concluded: “…’news deserts’ are fairly prominent across the sample, with eight of the communities containing no news stories, eight containing no stories addressing critical information needs, 12 containing no original stories, and 20 containing no local stories.”

However, the study was restrictive in how it defined “local,” and it gave every site equal attention and value, regardless of its capacity to produce quality journalism. That methodology raises the question: Did the study’s methodology skew its conclusions about local news quality in a downward direction?

“Understanding Local Journalism” did not render judgments about specific news sites. So I went to Bill Church, senior vice president of news at GateHouse Mediathe biggest publisher of newspapers in the U.S.—with questions about the quality of the GateHouse sites that were put under the microscope in the 100-community Duke study.

Five websites from GateHouse newspapers were examined in the study. What’s your assessment of those sites?
We reviewed markets that included GateHouse Media properties and found plenty of recent examples of GateHouse coverage that reflect original stories of local interest. GateHouse newsrooms in the study markets and others of similar size (populations 20,000 to 300,000) typically average more than 30 news articles and photo galleries daily. The majority of editorial content in these posts is locally based.

What are examples of editorial coverage by GateHouse community sites that might have been assessed in the study?
Let’s start with the Spartanburg (S.C.) Herald-Journal ( A recent series on high-poverty schools prompted this response from two area superintendents to Herald-Journal Executive Editor Michael Smith: “We wanted to take a moment to say THANK YOU.… Most importantly, you have informed our community (here and beyond) of the challenges facing many of our children and their families. We have shared these articles with a number of groups, and we are optimistic they are having a positive impact on our work. Thank you … for ensuring the Herald-Journal remains a community partner.”

Here are recent examples from GateHouse news sites in other communities that were part of the study:

The Beverly Homecoming Festival, a tradition for more thab 50 years, is well covered by your  website in the Boston suburb, but it does not appear to meet the Duke study’s criteria for “critical information needs.” Those needs are identified as “1. Emergencies and risks, 2. Health, 3. Education, 4. Transportation systems, 5. Environment and planning, 6. Economic development, 7. Civic information, 8. Political life.” Your view?
For the sake of discussion, it wouldn’t matter if the Homecoming Festival had been categorized under the Duke study’s mystical Category No. 9. An event that has drawn thousands of people annually for half a century fits the definition of what’s local and what’s important in Beverly. And my sense is that it blends several categories such as “economic development,” “civic information,” and “political life” if the coverage is looked at organically from its inception. Traditions and storytelling legacy factor heavily into a community’s brand, and local media understand that connection.

We’re not quibbling with the study’s methodology. All researchers have to pick a data path. Methodology matters for both academic research and data-based reporting. The Duke researchers should be lauded for tackling a question that reflects daily newsroom discourse on what is locally relevant.

There are 14 communities in the Spartanburg Herald-Journal-GoUpstate’s coverage area. They are Spartanburg County and the City of Spartanburg and two other cities and 10 towns—and they all have their own governments serving a grand total of just under 300,000 people. There is also the self-funded, 11-county Spartanburg Regional Healthcare System, which does 130,000 surgical procedures annually. You cited your Spartanburg’s website’s coverage of high-poverty schools. How effectively does the site cover the quality and cost of the services provided by the 14 local governments and the regional health-care system, all of whose budgets total in the tens of billions of dollars?
Michael Smith, our editor at the Herald-Journal, would respond in the same way as many of our top editors: We cover our communities the best way we can with our resources.

Our four newsrooms in the western Carolinas are a good example of figuring out how to share resources while doing journalism with impact. Does every governmental entity get the same level of coverage? No. But you could make the argument that’s been an ongoing challenge for decades.

I’m not sure how the Duke researchers dealt with local geography and the complex, highly nuanced labyrinth of neighborhoods in each community, so who knows if staff coverage of other cities and towns would be credited. The question may be: Does getting “credit” matter to the local subscriber?

How does GateHouse determine what its papers, which include websites for its 145 dailies and 340 weeklies, will cover, based on particular needs, and how does it measure performance? How do the papers score against standards you set?
We ask our local newspaper leaders to pay attention to high-interest, high-engagement news coverage and match resources with coverage priorities. There is no mandate on what they must publish or cover locally.

In the past two years, GateHouse Media has invested in a news innovation lab and national data projects team to partner with local journalism initiatives. We’re rolling out national investigative reports with local connections while also supporting small newsrooms such as The Hawk Eye in Burlington, Iowa, with its top-tier work.

Our innovation team—known as de//space—has developed an audience-reach dashboard that allows us to compare newsroom performance in multiple ways. It’s an internal-only measurement still in the pilot stage, but we’re excited about its potential to give quick, deep summaries.

Our newsrooms are assessed quarterly through the Inner Circle program, which looks at a newsroom’s local and enterprise/investigative reporting, overall planning, and digital acumen. Newsrooms are categorized and measured by staff size.

The Duke study rated the news of local operations in its 100 randomly selected communities against three criteria—how local and original was the news and did it meet “critical information needs.” Are there examples when coverage may be “local” but not counted that way in the study? If so, do you think what’s “local” should be flexibly defined community to community?
Definitely. On any Saturday during the college football season, you’ll find more than six different “team flags” planted in garden beds in our suburban Austin neighborhood. It’s a telling example of the evolving puzzle matching local passions with local news products. The Austin American-Statesman does a stellar job of covering the Texas Longhorns, but the garden beds in our subscriber-friendly neighborhood indicate just as much interest in the Ohio State Buckeyes and, yes, Oklahoma Sooners.

The Duke study put well-intended emphasis on reporting that is local, original, and which addresses a critical information need. But a regional story on water rights in southern Colorado or a report on hog farm controversies in rural North Carolina may draw more discussion and interest at the local McDonald’s and coffee shops.

Local editors, particularly in rural markets, also will tell you that readers are more likely to discuss national political coverage than what happened at the recent planning board hearing. Your version of local news isn’t your neighbor’s, and that’s the reality.

Did the Duke researchers go to GateHouse with any questions about its five news sites that were in its 100-community study?
We’re not aware of any questions the researchers brought to GateHouse.

As senior vice president of news at GateHouse, what is the likely next major step you see your operation taking to improve local news coverage, either overall or selectively by community?
We’re in the proof-of-concept stage now, working with newsrooms in several states on using design- and agile-thinking to improve the local news experience. More to come, but we’re addressing the realities of news resources and evolving audience dynamics.

Do you think the investment in improvements in the quality of local news will pay dividends, especially in reader conversion to subscriptions?
Yes. It’s happening now, thanks to the company’s investments in digital audience growth and consumer marketing.

Tom GrubisichTom Grubisich (@TomGrubisich) has written “The New News” column for Street Fight since 2011. He is also working on a book about the history, present, and future of Charleston, S.C.