Local Journalist Index 2025

Background

This project started during a conversation between Steve Waldman, founder of Rebuild Local News, and Scott Yates, an early supporter of RLN and member of its advisory board. Waldman noted that a community’s quantity of local journalists is as important as its quantity of news outlets — but that we had little information about how many journalists are on the ground in each place. A team led by professor Philip Napoli of Duke University had come up with estimates in 2018, but those were only on the state level and proved impossible to replicate each year.

Could we devise such a number that would get down to the county level — and in a way that could be repeated each year? That would be useful for philanthropists (to understand where the need is greatest and whether interventions are working), for public policymakers (to see where policies should be targeted) and maybe even for entrepreneurs (looking for opportunities).

Yates realized the answer may lie in partnering with a public relations software company. In part what those companies do is sell software and services to firms trying to reach journalists, so it’s in their interest to have a good and accurate database of journalists. Rebuild Local News teamed up with Muck Rack, a software platform widely used in both the journalism and public relations sectors. Muck Rack was originally founded as a tool for journalists to connect on social media and has since evolved into a platform used by communications professionals and journalists.

Uniquely, Muck Rack maintains comprehensive and current media data by combining automated data collection with the work of in-house editors. It is currently the only platform of its type with an editorial team that manually vets and verifies both journalists and outlets. Journalists can also claim and update their own profiles, introducing a layer of self-verification that enhances data accuracy.

The system indexes hundreds of millions of data points related to journalists and media organizations across digital, print, broadcast, podcasts, newsletters and social media. In total, Muck Rack monitors over 600,000 global news sources and more than 3.5 million articles daily. Its profiling system identifies individuals responsible for producing or distributing news content, assigning authorship and compiling metadata such as outlet affiliation, publishing frequency and geographic signals.

Methodology

For this analysis, we looked at millions of articles and sources from January 1, 2025 to March 31, 2025. This report drew on two primary components of the Muck Rack dataset: first, a county-level aggregation of journalist presence across all 3,141 U.S. counties, and, second, a record of more than 100,000 individual contributors whose work was analyzed according to factors including output volume, outlet type and local relevance. This included all topics, including sports, opinion and commentary. This also was not limited to the English language; all outlets based in the U.S. that Muck Rack tracks were included, like this one, for example. The data often does include photographers, broadcast journalists and others, as long as their work appeared online with a byline.

The data used in this particular study is exclusively focused on articles that have been published online. That leads to certain gaps in our knowledge.

A number of small outlets are still print-only. If they don't have a website, they are not included. Some newspapers use social media as their only online presence; as of now, they are not included.

TV and radio pieces that are broadcast on air without also having a version posted online sometimes are not counted. Some small outlets have antiquated and/or irregularly maintained content management systems that make their stories nearly impossible to ingest into an online system.

Some papers have paywalls constructed in a way that blocks content searches. Those have not been counted. Others have paywalls that allow for analysis of headlines, snippets of text, and bylines; those are counted.

Some articles don’t have bylines and some journalists don’t write articles with bylines. Editors, for example, often do their jobs behind the scenes without publishing bylined pieces. Some broadcast journalists don’t publish their work online with their names attached. This spreadsheet essentially excludes anyone who didn’t publish any stories with bylines online in the first quarter of 2025. (The formula does add 10 percent of an LJE to the county if that person has an active profile in the Muck Rack data and is linked to a local outlet, but didn’t publish in the first quarter of this year. That gives a small toe-hold to include editors and others who have been included in Muck Rack’s data, but didn’t have any bylined articles.)

Some outlets use a catch-all byline, like “Editor,” that can’t be assigned to an individual because it’s impossible to know if that is one or more people. They are not counted. Some people cover more than one county, often in a group of sister publications. We statistically assigned that person to one county, so some counties may show zero journalists even though those counties get regular coverage from a portion of a journalist's time. This may, in some cases, create an overstatement of coverage for the county that the journalist is placed in and an understatement for the other counties that the journalist covers.

We placed journalists in counties by associating them with the outlet that published their work. If a journalist published in more than one outlet, we used the one that was most common during the first quarter of 2025. The location of each outlet is typically confirmed by Muck Rack’s research team.

Creating the Local Journalist Equivalent

These datasets provided a gold mine of valuable information. One section has a field for each of the 3,141 counties (and county equivalents) in the nation. The other section has a field for each of the approximately 100,000 people in the data provided by Muck Rack. But it needed to be refined. In some ways it overstated the number of journalists; in other ways, it undercounted. So we made adjustments.

First, we narrowed down the list.

Muck Rack maintains profiles of people who “are responsible for the consistent production and/or distribution of newsworthy content,” which casts a wider net than we needed for this project.

Then we needed to assess which of these journalists were doing mostly local coverage.

We created a rough count of journalists in each county — by limiting it to people whose work was mostly published in that county.

But this still included a large number of journalists who weren’t doing local journalism. This included anyone focused on national topics — music, lifestyle, national politics.

So we added more filters:

Freelancers: We didn’t want part-time freelancers to be counted as full-time local journalists because that would overstate the amount of coverage provided. On the other hand, a part-time freelancer does provide coverage, so freelancers count as a smaller percentage of an LJE in the overall number, depending on how many stories they produced and in part also on the other filters.

Number of stories: If someone published one story in the first quarter of 2025, the “publishing” multiplier for that person was 0.1, a number that was picked after experimenting with dozens of different variations of formulas and starting points and testing that against real-world journalists, counties, outlets and populations. From there, we used a sliding scale that went up from 0.1 based in part on the journalist’s output in the first quarter of 2025 and in part on lifetime stories published. So if a person was recognized by Muck Rack as contributing to at least 20 stories between Jan. 1 and March 31 of 2025, they would be counted as one full LJE as long as there were no other multipliers (like the “aggregator” tag) applied to that person.

Aggregators: We filtered out journalists whose bylines commonly appeared on sites that Muck Rack had determined to be aggregators of content. These are generally services that provide syndicated national content that is not associated with an individual location.

Local: The unfiltered database included many national journalists as well as some public policy firms, think tanks and others that created content they tailored for local publications but produced elsewhere. We needed to develop a scalable approach to focus on local journalists. Someone was deemed to be a local journalist through a multi-step process.

First, we associated journalists with outlets covering local news in a town, city or county. Muck Rack confirmed that the outlet wasn’t just located in a particular place but actually covered the community.

If a journalist's work appeared in more than one outlet, we "assigned" them to the one they were published in the most often (in the first quarter of 2025).

Finally, we applied an "aggregation filter" to make sure they weren't publishing articles all over the country in a way that indicated that the stories weren't really local. Simply put, we filtered out journalists whose content was appearing commonly in multiple locations, often 100 or more outlets. We did this using two different sliding scales for the number of outlets in the first quarter of 2025 and the number of outlets lifetime for each individual journalist. Those formulas also filtered out journalists whose work appears as a part of a wire service, large national outlets or syndicated columns.

Coverage from a nearby metro area: Just because there are no journalists in a county, that doesn’t mean there is no local coverage. For example, in southwestern Colorado, you’ll find Dolores County. Muck Rack counted no journalists, and the most recent Medill report on news deserts shows no outlets there. But if something newsworthy happens in Dolores County, a big-city outlet might cover it. For example, one TV station covered a rock falling on the road.

All of that is to say, the journalists listed in Denver cover Denver, but they also cover the suburbs of Denver, and they cover the whole state, too, which is typical of every state.

So we created a formula to represent that.

The formula takes the number of Local Journalist Equivalents associated with the largest city in a metro area and assigns 60.16% of that number to the county containing that core city. Then, 26.44% is assigned to the suburbs of that core city. The remaining 13.4% is then divided among all counties in the state except the county containing the core city. Suburbs get a sliver of that number depending on the other factors in that state, so they sometimes end up with a hair more than 26.44%.

Where did we get those numbers? We obtained information about 28,000 stories from a collection of outlets in Denver, including the Denver Post, the Colorado Sun, Colorado Public Radio, five television stations and one radio station. We picked Colorado because it is pretty average in terms of urban/suburban/rural breakdown, and because Yates is from Colorado and was able to analyze the data and see problems based on his knowledge of the media markets and the cities and towns.

One artifact of that metro area adjustment is that some counties and states have numbers that may seem to not add up. This happens when a metro area crosses state lines.

The database contained metadata for each story and a snippet from the story, typically the first paragraph or up to the first 100 words. We then used a lookup function and compared those snippets against a list of cities and towns in Colorado. (We had to create an additional check for town names that appear in news stories in other contexts, such as the town of Center.) Then we counted the stories that had the word Denver first, then cities in the suburban counties and then cities and towns in the rest of the counties. There were 3,672 for Denver, 1,614 for the suburbs and 818 for outlying cities and towns. (By the way, that adds up to 6,104 stories, meaning only about 20% of the news published by those local outlets in the first quarter of 2025 was identifiably “local.”)

After making these adjustments, the pool of journalists dropped from about 100,000 down to about 27,000 Local Journalist Equivalents.

How do these numbers compare to other estimates of journalists?

Probably the most accurate and matching number from 1992-2002 is the number of 116,000 from research led by David H. Weaver at the Indiana University. (Now emeritus, he provided helpful guidance to us in creating this report.) The Indiana team created that estimate (visible in this later paper) by drawing random samples of different kinds of news organizations (daily and weekly newspapers, magazines, radio, TV, wire services, news websites) and then randomly sampling the lists of journalists that they obtained from these news organizations. They used directories from news organizations to draw the first sample and then would estimate how many journalists worked for each type by using the lists. Our number is different because it includes freelancers, but it doesn’t include people whose work isn’t published online. There’s no perfect comparison, but we concluded this is the most accurate snapshot of that period, and close to an apples-to-apples point of reference, an assessment that Weaver also found reasonable.

It is also in line with other studies and data from the Bureau of Labor Statistics, which show a decline in the same range. And our current data is also similar overall with the most current BLS data, even if it has some big differences in some individual states. The most recent BLS report is from May of 2023, so nearly two years before our sample of data, so doesn’t include nearly two years of continued contraction in journalism employment. In that BLS report, we compared our data state by state with that data, and our numbers show an average of 356 fewer journalists per state. New York is understandably the most divergent as it has so many journalists not working in local news. The BLS data showed 5,960 journalists working there as of May 2023. In the original data from Muck Rack we had more than 8,000 people, but after filters described above, that number dropped to 1,912 LJEs. That was the state with the biggest difference. Our numbers were almost identical to the BLS in many states. For example in Missouri we calculated 596 LJEs, and the BLS reported 580 journalists.

We plan on making this an annual report, so going forward, we’ll have better trend data with direct comparisons.

Reality testing against particular localities

Getting to our number of Local Journalist Equivalents was an iterative process. The first batch of data only included people who met a certain threshold of data quality within Muck Rack’s database. That left out many legitimate journalists working at small publications with, perhaps, outdated websites or a lack of other external verifications that are common for large news outlets but less common for outlets that serve smaller communities. The second batch turned off all of those data quality filters, but then it included people who wrote nonlocal stories that appeared in local publications. The third batch of data included all those details in the list above, and we were able to tweak the formulas to get to the most accurate version of the Local Journalist Equivalent.

There are some oddities. For example, the Ohio county with the most local journalists per resident is Athens County, but it is helped by the presence of Ohio University. That’s the case for a few other counties with big universities, such as Albemarle County, home of the University of Virginia in Charlottesville. Those two universities have great student journalists who show up in strong numbers in the Muck Rack data and therefore in this report. The data doesn’t show the same phenomenon in every county that’s home to a robust student news outlet, probably for reasons related to the content management system, byline policy or other factors.

Working with county-level data comes with a number of challenges. Connecticut doesn’t technically have a county government system, and it recently changed the number of the geographic regions it calls counties. For logistical reasons, in this year’s report, we are using the old system of counties for Connecticut. Virginia also presents a challenge because some of its cities are independent from counties, meaning they’re not legally part of any county even if they share a name. For example, Fairfax County and the independent city of Fairfax are separate entities. Similarly, Baltimore County in Maryland abuts the city of Baltimore but does not include it. To account for this, we grouped independent cities with their surrounding counties into metro areas for the purpose of allocating journalists. And yes, we’re aware that Louisiana has parishes and Alaska has boroughs and census areas, but on the map and in the database, they are all counties.

On top of all that, the process was not easy for reasons that often have to do with the crisis in local news. We found small counties that had what looked like a news site at a domain that had once been a news site, but the owners of the paper — perhaps inadvertently — let the registration of the domain slip away and now the “news” there is nothing but crypto scams. Some small papers have only a static page with the locations where people can purchase the physical paper. Some radio stations report local news on the air and never publish any of the stories online. Those are, sadly, not included in this data as we wanted to be rigorous and consistent in using the available data.

Other challenges cropped up because, in a few cases, small papers fill their sites with stories written by think tanks or advocacy organizations. That’s not local news. A news site may run a piece about, say, gun rights and tailor it so that it mentions the name of the town a few times, but the story has no actual connection to the community. We had to tweak our searches and algorithms over and over to attempt to keep those think tanks out of the count of local journalists while not excluding freelancers and stringers who are actually in communities reporting about local events but are not full-time staffers.

How did we compare the results among rural, suburban and urban areas?

While looking at either the county map or the state map, it’s clear that the shortage of journalists is widespread and is based in every kind of community no matter the size, geography, or population. To double-check that, we ran two tests on the underlying data. This chart shows the count of Local Journalist Equivalents and the population of counties in a set of five bands. Here is that result:

This analysis provides a rigorously constructed and independently reviewed snapshot of the state of local journalism in the United States as of the first quarter of 2025. It is based on the most comprehensive and current dataset of U.S. journalists, maintained by Muck Rack and regularly updated by journalists themselves. The methodology — developed by Yates in collaboration with Rebuild Local News and Muck Rack — has been vetted by academic and industry experts (see Acknowledgements). While no dataset can capture every nuance, this work establishes a new benchmark for understanding the reach and distribution of local journalism nationwide. We hope it will be a useful tool for all those who are working toward the goal of strengthening local news because it is such a vital part of a healthy democracy.

What comes next?

We welcome opportunities to collaborate with others to help shed light on the importance of local news. Areas for exploration include:

• What is the political breakdown of these journalist shortage areas?
• What is the relationship between the ownership structure and the likelihood that a state or county will have journalist shortages?
• What’s been the impact of nonprofit newsrooms or digital hyper-local startups?
• What impact do journalist shortages have on the economic health of a community?
• How could we include analysis of the content — and its quality — into future studies?
• What accounts for the huge differences between states?

In-depth map

In this map, when you hover over a county, you will find the following numbers, which derive from the methodology explained above:

1. Original. Muck Rack provided a list of journalists — national and local, freelance and full time. Our database formulas searched through that list and came up with a number of journalists that we could link to an individual county. The number you see here is the count of all the people we could confidently place as producing stories in this county from the provided list.
2. Work Adjusted. The number you see here is what we get when we narrow down the “1. Original” list by using various filters. We put those filters in place to see if the journalist is a freelancer, appears in aggregators or didn’t publish in the first quarter of 2025, along with other factors. Depending on what we found, an individual journalist may appear in the data as one journalist equivalent, zero, or a fractional value between zero and one. The number you see here is the sum of all those journalist equivalents in this county. This typically reduces the original number by half or more.
3. Metro Adjusted (LJE). After we have the “2. Work Adjusted” number, we then add (or subtract) journalists proportionally in each county based on a formula that spreads some of those journalists from the core city of a metro area to the suburbs of that core city. The formulas also proportionally shift some of the journalist count from the biggest city in the state to all the outlying counties in that state. This number therefore shows a reduction in the count for core cities but shows an increase for suburbs and rural counties. This is the last adjustment made to the count of journalists for each county, and it results in the Local Journalist Equivalent figures used throughout this report.
4. Population. The population of the county.
5. Original per 100,000 people. The original number (1) per 100,000 residents.
6. Work Adjusted per 100,000 people. The adjusted number (2) per 100,000 residents.
7. Local Journalist Equivalent (LJE) per 100,000 people. The number after all the adjustments (3) per 100,000 residents. (This is the "LJEs per 100k" figure that is in the box from the main map and is referenced throughout the report.)

Acknowledgements

This project builds on the insights, leadership and generosity of many people who have worked for years to understand and improve the state of local journalism in the United States.

We are especially grateful to the team at the Medill School of Journalism at Northwestern University, whose groundbreaking State of Local News reports have been essential in raising national awareness about the loss of local newsrooms and the communities most affected. This study is complementary. The research started by Penny Abernathy, now continued by Tim Franklin, Zachary Metzger and the Medill Local News Initiative, has shaped how policymakers, philanthropists and journalists themselves understand the landscape of local news. Metzger provided a very helpful review of the methodology, even though we couldn’t follow his suggestion to update the definition of counties in Connecticut (which recently changed county lines) because it would have broken several other elements of our project. Next year! And Abernathy gave sage advice that only someone with her long history with this material could have provided.

Philip Napoli paved the way for this approach with his pioneering work on journalists-per-capita metrics. He and associates Asa Royal, Sofia Bliss-Carrascosa and the rest of the team at the DeWitt Wallace Center for Media & Democracy at Duke University have done leading scholarship around media equity, coverage gaps and topic-based content analysis (e.g. coverage of local government, education and public health). In addition, Michelle Ferrier also did important work on “media deserts.”

David H. Weaver, professor emeritus at Indiana University, helped lay the groundwork for understanding the scope and structure of the journalism profession in the U.S. His co-authored volume, “The American Journalist in the 21st Century” (2007), provided a detailed snapshot of the profession at the turn of the millennium, including the figure we cited of 116,000 full-time journalists in the year 2002 — a benchmark we used to calibrate long-term trends. We’re grateful for the time he took to respond personally to questions and share helpful context about the methodology of that research.

Jessica Mahone, the interim director of the Center for Innovation and Sustainability in Local Media at the University of North Carolina Hussman School of Journalism and Media, have continued to work on these issues and made time for us at the excellent convening of local news researchers at UNC in March of this year.

Also generously providing time and expertise included: Damon Kiesow at the University of Missouri and Sarah Stonbely at the Tow Center for Digital Journalism at Columbia University; the American Journalism Project team, especially Loretta Chao; and Meredith D. Clark, a professor at UNC Chapel Hill and a former local journalist herself.

We also want to acknowledge the important work of other state-level news mapping efforts that have helped surface local coverage gaps and sparked public awareness and policy conversations. These include the Colorado News Mapping Project housed by the Colorado News Collaborative (COLab), the New Jersey News Ecosystem Mapping Project developed by the Center for Cooperative Media at Montclair State University, and similar projects in North Carolina, California and elsewhere. Their localized, collaborative approaches informed the structure and goals of this national-level effort, and we hope our work builds on and complements theirs.

We especially thank the board, the steering committee and the advisory board of Rebuild Local News for its guidance and feedback on the methodology for this report. Members of the board — including experienced journalists, nonprofit leaders, advocates and researchers — offered invaluable feedback on early drafts of this work and helped ensure that our assumptions, data sources and framing aligned not only with reality but also with the needs of practitioners in the field.

Finally, we thank all the journalists who are a part of this dataset. Here, they are only digits in a spreadsheet, but they are real people and are quite often doing heroic work under difficult circumstances. Many of them work without much recognition, compensation or institutional support. This project is ultimately about them and the communities they serve.

About the Authors

Muck Rack was founded in 2009 to connect journalists on social media. In 2011, the Muck Rack software platform was born, offering journalist profiles and a media database that gained a reputation for being the most accurate in the public relations industry. Since then, it has grown into a comprehensive PR software that nearly 6,000 organizations worldwide rely on to build trust, tell their stories and demonstrate the unique value of earned media. Muck Rack continues to invest in free journalist solutions, research and media data, underscoring its commitment to serving journalists and supporting a free press. Today, thousands of journalists use Muck Rack’s free tools to showcase their portfolios, track coverage, analyze news trends and measure the impact of their stories. This report was produced by the Muck Rack marketing team, including Mike Schneider, VP of brand, and Matt Albasi, data journalist.

Rebuild Local News is the leading nonpartisan, nonprofit coalition developing and advancing effective public policies designed to strengthen community news and information. Collectively, we represent more than 3,000 newsrooms and over 15,000 journalists, working together to safeguard the future of local news.

Scott Yates is the founder of JournalList.net, the caretaker of the trust.txt standard, and a member of the Rebuild Local News advisory board. He is also a strategic consultant who works on complex projects (like this one). Before that, he was a startup founder. Before all that, he was a reporter in New York City and his home state of Colorado.