Step 6: Investigate Through Analysis
You have now taken the necessary steps ahead of conducting your analysis. You have established your research purpose(s) and selected your unit(s) of analysis. You have defined your geographic scope, your timeframe, and your key concepts. You have determined which approaches and datasets are most relevant to your needs and can answer your research questions. The next step is conducting your analysis. This will look different depending on what route you wish to take.
Ideally, you will combine approaches and datasets to obtain the most thorough and holistic picture of journalism employment in your given geographic region and timeframe. This might look like a combination of a state-level survey with desk research or content- and byline-based approaches. In other cases, depending on needs and resource constraints, you might rely solely on desk research or federal statistical data. Regardless of the approach you ultimately choose, remember your research purpose and the audiences you seek to reach.
Putting Research Principles into Practice: A Washington case study
The most robust way of documenting journalism employment in your given geographic area and during your specific timeframe likely results from a combination of approaches. To put these approaches into practice, we have included a case study from Washington state, From News Deserts to Nonprofit Resilience: Assessing the Health of Washington’s Local News Ecosystem.
A team of researchers led by Washington State University Assistant Professor Dr. Jennifer R. Henrichsen carried out a local news ecosystem analysis of Washington. The purpose of the study was to understand the health of the local news ecosystem in Washington to inform policymakers, foundations, journalists, and academics. The project began because the Washington State Legislature appropriated $2.4 million to Edward R. Murrow College of Communication at Washington State University to create the Murrow News Fellowship program. As part of this program, Murrow College thought it was important to understand the local news ecosystem of Washington to ensure journalists were assigned to the most underserved areas in the state.
The unit of analysis for this study was the news organization as researchers were most interested in understanding how many news organizations existed across the state and where gaps in news were most prevalent. The research team first defined key concepts including what constituted a local news organization. To be considered local, outlets had to either be a) located in Washington and primarily oriented to a Washington community, or b) located outside of Washington but primarily oriented to a Washington community. Additionally, to ensure that the news organization produced local news, the research team only included organizations that a) publish substantial content on a website or social media page that is b) original, c) journalistic, and d) locally relevant.
The research team began collecting open-source data in fall 2023, interviewed 32 journalists, policymakers, and academics and disseminated a survey to the 353 outlets that met the criteria for database inclusion in 2024.
The research team sought feedback from other scholars in the field, especially those who had conducted or were in the process of conducting their own state-level ecosystem analyses. The lead researcher also presented the research at academic and practitioner focused conferences for additional feedback.
While the focus of the study was an overall mapping of the local news ecosystem of Washington, one metric the research team was interested in was journalism employment. Since several metrics were difficult to obtain via open source methods, including staffing data, the research team designed a survey questionnaire to get a more accurate and granulated count of journalism employment. In the Washington survey, top management officials (e.g., editors, publishers, owners) were asked the following questions as it pertained to staffing in their organization:
Q.7: What is your organization’s staff size?
|
Sample question from the From News Deserts to Nonprofit Resilience: Assessing the Health of Washington’s Local News Ecosystem. View Appendix A for more questions. |
One major finding from this question was how precarious journalism employment is at a majority of outlets across the state. Out of the 144 news outlets that provided full-time editorial staff counts via the survey, more than half (54%) have two or fewer full-time editorial employees.
Such numbers underscore the precarity facing the majority of news organizations in Washington state and align with other research from INN and LION that reveals precarity in local news organizations across the nation.
Responses to this question also illustrate how decisions about including part-time staff can significantly affect overall headcounts. Part-time editorial staff, including freelancers and volunteers, accounted for slightly more than half (53%) of the editorial staff total from all responding outlets. Below is a figure showing a selection of larger counties in Washington with higher survey response rates. The figure reveals how news organizations rely on part-time editorial employees and student media to carry out work at their news organizations.
Below, you can see how the data at the individual journalist level compares from the Muck Rack data to the Washington state survey at the county level. In essence, the Muck Rack data drew on a county-level aggregation of all counties across the US and an analysis of 100,000 contributors work according to output volume, type and locality1, while the Washington data included full-time and part-time editorial staff employed at qualifying news organizations (per the inclusion criteria previously discussed), according to survey results from managers, editors, and publishers. It should be noted that some survey respondents interpreted “editorial staff” differently or were more thorough in their counts. This led to a few instances where small outlets that diligently reported all part-time contributors appeared to have greater staffing than large newsrooms that reported only full-time journalists directly involved in original reporting. Another issue that arose was double-counting in some counties, such as when two outlets in rural counties share the same three employees. A comparative glance shows distinctions in coverage across the two data generating processes. Despite variation across smaller counties, the survey method utilizes a finer lens to understand the distribution of journalistic labor across outlets. This methodological approach provides greater granularity of journalism employment in news organizations than other options, such as BLS data. To see how news organizations in counties responded to the survey, please see Appendix C. [cut this: specific counties responded , which comes through as near-consistently better coverage even across which may otherwise be considered news-poor. ]
(Note: The difference in counts is not because of inaccuracies, but because of the way each data stream measures the object in question.)
The Washington research team is interested in conducting additional research, including analyzing the number of news outlets and their newsroom staff and how these vary by county and by region. The research team is also interested in evaluating how news ownership type affects news production capacity (operationalized as staffing, budget, and publication frequency). To carry out these future steps, the team hopes to explore a content and byline approach to measure newsroom content volume and similarity. Additionally, depending on funding, the team intends to conduct a second survey to assess how the ecosystem, including staffing, may have changed since data collection finished in 2024.
- Muck Rack. “Local Journalist Index 2025: Background & Methodology,” accessed May 4, 2026, https://muckrack.com/research/lji-methodology#methodology ↩︎
