Expert Commentary
Two reporters explain how they overcame major barriers investigating FEMA and share tips on interviewing hesitant sources, building data journalism skills and more.
It took a lot of work to get the numbers the Federal Emergency Management Agency would not share. After Washington Post reporters Hannah Dreier and Andrew Ba Tran figured out how to scrape the data from a dormant government website — pulling 9.5 million records, 1,000 rows at a time — they discovered how badly FEMA had failed the people it was created to help.
The federal agency’s mission is to prepare the country for and respond to natural disasters such as hurricanes, wildfires and floods. FEMA also oversees federal recovery efforts, including distributing funding to help communities rebuild.
Dreier and Tran show that FEMA has denied the overwhelming majority of disaster survivors’ requests for aid in recent years. At the time the two journalists reported their findings in 2021, FEMA rejected nearly 90% of applications seeking help repairing homes, burying loved ones and covering various other costs of recovery. Even as the need for this type of assistance rose nationally, they learned that application approval rates fell from 63% in 2010 to 13% in 2021.
Throughout 2021, Dreier and Tran investigated FEMA and its work going back years. After traveling the country, conducting close to 300 interviews, submitting dozens of public record requests, analyzing thousands of documents and creating multiple databases, they published a series of stories that also showed how FEMA:
The series initiated change, prompting new legislation and forcing FEMA to revamp its policies, publish application approval rates, and partner with the U.S. Department of Housing and Urban Development on housing assistance and case management.
In December, President Joe Biden signed an executive order directing the Department of Homeland Security, which oversees FEMA, to “design and deliver a streamlined, online disaster assistance application” and work with individual states to make it easier for people of all ability levels to apply for and receive disaster aid.
The Journalist’s Resource reached out to Dreier and Tran to ask them about their work and for tips to guide other journalists hoping to pursue similar projects. Eager to help, they offered insights and advice across a slew of subjects. Below, we share six of their tips on framing stories, gaining access to government data, interviewing and building relationships with sources, and learning how to do data journalism.
Because FEMA declined most requests for interviews and information, Dreier and Tran found other ways to get the information they needed to understand how the agency responded to disasters and requests for aid. Dreier, a national enterprise reporter at the Washington Post, traveled the country visiting ravaged communities and seeking interviews with disaster survivors and FEMA workers.
Getting strangers to talk about their personal struggles and employees to divulge information an employer does not want a news outlet to have can be quite difficult, Dreier says. But she keeps trying to reach people and interview them until they make it clear they are uninterested.
“I try to assume everybody wants to talk to me — until they tell me directly that they don’t,” says Dreier.
FEMA officials had insisted employees working at the call center for its COVID-19 funeral assistance program would not speak to her. When Dreier asked advocates for California families who lost their homes in wildfires for help contacting them, the advocates told her families did not want to talk.
They did, though. And Dreier would not have gotten those interviews had she not visited the FEMA camp in Chico, California and asked residents to share their experiences or used LinkedIn to track down close to 100 call center employees. One worker even allowed Dreier to spend the day with her as she answered phone calls seeking aid from her home office.
“I just think it’s important to keep going until you really know the person isn’t interested in talking,” Dreier says. “I flew out to California and showed up on the doorstep and they were super happy to see me. They welcomed me right in and were glad someone wanted to tell their story.”
“Don’t always look for bad guys,” says Dreier, who tried to understand the various factors influencing FEMA decisions.
She doubts agency officials acted out of malice when they denied people’s requests for assistance. But they also had not made changes after realizing certain rules prevented qualified applicants from getting aid.
Dreier, who did the bulk of the writing for the project, drew upon interviews and observations of FEMA employees to demonstrate how rules designed to help people can have unintended consequences. For instance, FEMA’s policy on which documents it accepts as proof of ownership of a property has made it impossible for some Black families to qualify for assistance rebuilding after a disaster.
More than one-third of the land Black people own in the southern U.S. “is passed down informally, rather than through deeds and wills,” Dreier and Tran write. “It’s a custom that dates to the Jim Crow era, when Black people were excluded from the Southern legal system. When land is handed down like this, it becomes heirs’ property, a form of ownership in which families hold property collectively, without clear title.”
In the second article in the series, Dreier describes how a FEMA specialist tried unsuccessfully to help Black families who had lived for generations on the same property in Alabama get money to rebuild their homes following a tornado.
“The more you can help readers get into the heads of everyone in the story, the more likely they’re going to read the story and you can get at the different complexities,” Dreier explains, adding that the second story “showed that [FEMA] workers themselves were frustrated and helped drive home the point that these policies were really problematic.”
It’s important for sources to know that even though your reporting could benefit them and even though they may have shared personal details and spent considerable time with you, you’re not their friend.
Dreier recommends reminding sources of these things, especially sources who don’t have experience working with journalists. They must understand that anything they say and do — unless you have agreed to allow some of it to be off the record — could show up in a news story.
Two signs that sources might be confused about your role: They ask you for advice on how to handle the problem you’re reporting on, or they share information that could hurt them.
“It’s hard to initially build trust and then, once you have that trust, there’s always the worry, you don’t want them to trust you too much,” Dreier notes, adding: “There’s something about a person coming in and being so interested in the minutiae of your life that can make someone let down their guard.”
Writing in a notebook or using a voice recorder, regardless of whether you need to capture certain details or conversations, is a good way to remind people that you’re doing a job. Dreier says these visual cues “remind people not to give you access to information they’re not truly comfortable with.”
On occasion, though, she feels the need to explicitly state her relationship with a source.
“Sometimes, I’ll just say ‘I’m not your friend,’” she says. “It’s so awkward. I think sometimes you have to do that.”
When Tran, a data reporter on The Washington Post’s rapid-response investigative team, first tried to look at data on how often FEMA approves applications for aid, he could not get to it. The FEMA website had provided a link to the information but the link did not work. That meant Tran had to figure out another way to access the data.
He used his programming skills to find and pull data from the FEMA website’s software interface.
“It’s rare, but you can sometimes find it stashed in a corner of a website or in the background of the website,” Tran says.
“It’s just trial and error,” he adds. “You poke around and see what access points are available.”
He found the data on FEMA’s site and he and Dreier were able to scrape, or collect, the information piecemeal — 1,000 rows at a time.
He had to reformat the data by hand so he and Dreier could compare it against census tract data to see if there were racial disparities in application approval rates. There were.
“Nationally, FEMA denies requests for help from about 2 percent of applicants for disaster aid because of title issues,” they report. “In majority-Black counties, the rate is twice as high, according to a Washington Post analysis, in large part because Black people are twice as likely to pass down property informally. But in parts of the Deep South, FEMA has rejected up to a quarter of applicants because they can’t document ownership, according to the Post analysis.”
“Get the dataset they worked with or something similar that’s relevant to your area and then redo [the project],” Tran says. “You’ll learn something along the way, especially if these journalists are putting up their methodology on GitHub.”
GitHub is a social coding site and the world’s largest repository of source code. Many news outlets put their open source projects on GitHub, which allows journalists from other newsrooms to repurpose them for their audiences.
For instance, Tran posted on GitHub the data and analysis he did for the story showing that funding for FEMA assistance projects often doesn’t reach communities until years after the disaster occurred.
If you need help, Tran suggests reaching out to data journalists you know or whose work you follow. You can also find data journalists through the National Institute for Computer-Assisted Reporting, a project of Investigative Reporters and Editors, or the News Nerdery’s Slack channels.
“The data journalism community is so welcome and open,” he says. “You just have to find a project to apply these skills to.”
Tran says that’s how he learned. He completed several free online courses, including MIT’s 13-week course, “The Analytics Edge,” and an 18-hour course from Johns Hopkins University, “The Data Scientist’s Toolbox.”
He urges journalists not to let a lack of confidence in their math skills keep them from learning how to obtain and analyze datasets.
“I was horrible in math growing up — I was horrible,” he says. “Once you start getting into the habit of [analyzing data], you see the patterns and things. It’s not intimidating once you get past the first [data project].”
Tran has taught data journalism at Wesleyan University and American University. He also created a free online course, “Using R for Data Journalism,” to teach journalists how to use the open-source programming language R.
That course is part of a longer online course offered through the Knight Center for Journalism in the Americas at the University of Texas at Austin, “Intro to R for Journalists: How to Find Great Stories in Data.” Tran is the instructor for that free course as well.
Two experiments reveal that Democrats tend to see female and minority judges as less biased than white male judges. Republicans often hold the opposite view.
Learn what sanctions are, how they work and what sanction levers the U.S. and its allies have left to pull following Russia’s invasion of Ukraine.
Some politicians, public officials and a handful of physicians are spreading misinformation. It’s important for journalists to debunk falsehoods high up in their stories, avoid false balance, and dig into the motives. 
This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.
by Denise-Marie Ordway, The Journalist's Resource
March 25, 2022
This <a target="_blank" href="https://journalistsresource.org/politics-and-government/washington-post-fema-investigation-goldsmith/">article</a> first appeared on <a target="_blank" href="https://journalistsresource.org">The Journalist's Resource</a> and is republished here under a Creative Commons license.<img src="https://journalistsresource.org/wp-content/uploads/2020/11/cropped-jr-favicon-150×150.png" style="width:1em;height:1em;margin-left:10px;"><img id="republication-tracker-tool-source" src="https://journalistsresource.org/?republication-pixel=true&post=70457&ga=UA-5985794-2" style="width:1px;height:1px;">
Thank you for subscribing.
Harvard Kennedy School is committed to protecting your personal information. By completing this form, you agree to receive communications from The Journalist’s Resource and to allow HKS to store your data. HKS will never sell your email address or other information to a third party. All communications will include the opportunity to unsubscribe.
A project of Harvard Kennedy School’s Shorenstein Center and the Carnegie-Knight Initiative, The Journalist’s Resource curates, summarizes and contextualizes high-quality research on newsy public policy topics. We are supported by generous grants from the Carnegie Corporation of New York, the Robert Wood Johnson Foundation, The National Institute for Health Care Management (NIHCM) Foundation and individual contributors.
Find us:
Unless otherwise noted, this site and its contents – with the exception of photographs – are licensed under a Attribution-NoDerivatives 4.0 International (CC BY-ND 4.0) license. That means you are free to republish our content both online and in print, and we encourage you to do so via the “republish this article” button. We only ask that you follow a few basic guidelines.
