Investigative approaches

Students’ picks:

Thoughts arising from discussion

Think about the human sources who can help you

In some of the articles we discussed, the reporters may have been tipped off to the initial story by an interested party — who may also have provided documents or data. Some, notably the National Park Service sexual harrassment story, featured the stories of people who were wronged, and wanted to be given a public voice. At this stage in your career, you probably won’t have whistleblowers eager to hand you a story to pursue. But there are still plenty of people who will be motivated to help you. Academics and non-governmental organizations with an interest in the issue can be very useful, as can attorneys representing a party in a lawsuit. Of course, you need to be aware that most interested parties have an agenda, which you must bear in mind. But the information they can provide may still be very useful.

Former employees of government agencies and other bodies, freed from the constraints of “party line” or official silence, can also be good sources.

When approaching a source who may not emerge in a good light from your reporting, always think: Why would this person want to speak to me? Often, they will at least want their side of the story to be told. So frame your interview request in terms of giving them that opportunity, remembering that this should be a sincere offer. Also, go into your interviews with an open mind, and you may find that your view of the situation may shift once you’ve spoken to all of the interested parties.

To stress what will become a refrain from me: Hitting the phones and talking to as many informed sources as you can is crucial!

Many government datasets can simply be downloaded is the official repository for the federal government’s open data, but many of the most useful datasets are available through the websites of individual agencies. Martha will be guiding you though some of these resources in the coming weeks, but here are some of my go-to sources:

Enterprise database reporting

Government or other databases can be put to good journalistic uses, and most of the time, this need not involve sophisticated statistical analysis. The clinical trials reporting story, for example, was based on simple counts, calculations of rates, and so on. But note that visualization of such data should be simple to follow, not confusing. We’ll discuss good principles of visualization in week 5.

Note also that if you have a collection of documents, it may be possible to systematically extract information from them to create your own databases or spreadsheets, as I’m sure was done for the Wildlife Services investigation.

We’ll cover acquiring data from online, and the basics of journalistic data analysis, in the coming weeks. If you need help with analyzing data before then, do get in touch with me. Also, these resources from my website may be useful.

Court documents

These can be a treasure trove of information if the documents are not sealed from public view by the court. Search for federal cases at PACER (you will need to sign up for an account and pay a modest sum for any documents you download). Some states have centralized searches for their courts but in many cases you will have to go to individual courts to find the documents you need.

Tax records

Even if an organization has tax-exempt status as a non-profit/charity, it will still have to file an annual report to the Internal Revenue Service, which will contain lots of useful financial and other information. We will cover this in week 6.

Good sources of reports etc

Some of the stories we discussed mentioned reports from government agencies or other bodies. Here are some sources of reports that I find most useful:

Enforcement/inspection records

Agencies charged with enforcing laws or regulations will create a trail of documents that you can follow. Enforcement actions or even routine inspection reports are usually public documents that may be online and searchable, or can be obtained through public records requests. The Pulitzer-winning story on ship scrapping will have relied very heavily on Occupational Safety & Health Adminstration enforcement actions.

Meeting transcripts

If there was an official meeting, there will probably be minutes or a full transcript. These can be very useful documents.


If you need to find out how much a public project cost the information will usually be in government budgets or in contracts awarded. can be a good starting point for federal grants and contracts.

Public accountability versus personal privacy

This is often a tension that surrounds public-interest journalism. We mentioned HIPAA and the confidentiality of medical records. In some cases, personally identifying information may be redacted from public records. Other documents, such as death certificates, are public. Patients who are willing to talk about their experiences can often be found through patient advocacy groups, lawsuits etc.

Don’t forget the scientific literature

As science reporters, you should be well placed to gather all the published studies that are going to be relevant to your study. So as you explore other sources of data and documents, don’t forget to do scientific literature searches. Scientific papers can also be used in other ways, such as to determine sources of researchers’ funding from the acknowledgments section and conflict-of-interest statements.

Use your own expert knowledge

As you gain experience, you will become an “expert” on the beats you cover – in some cases you’ll enjoy an overview that other people closer to the specific parts of the action don’t have. Use this expertise, and combine it with a critical journalistic sense to start doing a little digging if things “don’t smell right,” or seem “too good to be true.” We’ll touch upon the last point in week 10, when we cover reporting on scientific misconduct.

Build on the work of others

With proper attribution, it’s perfectly OK to build on the work of other journalists. Other journalists’ FOIA requests or databases that they have compiled can be the starting point for your own stories. ProPublica, for instance, has encouraged reporters to work with its Dollars for Docs database and other data.

Even if your story is local, don’t think parochially

Always look for context and comparisons. Compare to the rest of the state, nationally, internationally, as appropriate. If you need to make comparisons between nations, the World Bank probably has what you need, its catalog containing data for more than 7,000 different measures, compiled by the bank and other UN agencies.

Financial disclosures/following the money

Financial disclosures can be very revealing, and are required in many circumstances. Here, here and here, for example, are some links on financial disclosure policy for University of California faculty. Martha will talk more about financial disclosures and following the money.

Know the difference between anecdote and data/analysis

I’m glad we had a frank discussion of the weaknesses of the Hawaii pesticide story in this regard!


Many activities require permits from federal, state or local governments, and these should be public documents. Despite its other weaknesses, the Hawaii story provided an example of using these documents.

Professional accreditation/licensing

Many professions are accredited or licensed by state government agencies. Doctors and state medical boards are the most obvious example, but there are many others. These records will usually be searchable online, and may include disciplinary actions. Here is California’s license search site, where you will find doctors and many other licensed professionals.

Trade/industry data

Some trade bodies collate information on their industries, and other organizations earn their living by collecting and analyzing industry data.

If appropriate, you can commission your own analysis

If you find an interested scientist, and the work isn’t enormously time-consuming or expensive to perform, this approach can be worth considering.

You may also be able to structure your interviews so that, in effect, you perform your own survey limited scope. The nail salon’s story provided an example of this.

But remember what I said in week 1 about running with scissors, and seeking expert advice. That’s particularly important with surveys, where proper methodology is crucial. This book from Philip Meyer, which I mentioned in week 1, is a very good resource.

Remember that the web is in flux

Web pages come and go, and can be changed at a moment’s notice. So if showing what a web page said may be crucial to your story, save it. I usually save as a PDF, using the Print options available in Firefox.

Even when web pages have disappeared, they may not be gone forever. Check out Google’s cache for the most recent version of a page crawled by its robots. Search here (replacing <Website URL> with the target page) or here.

The Internet Archive’s Wayback Machine is a great resource for finding old web pages, and looking at how sites have changed over time. It also has a “Save Page Now” option, which you can also use to save snapshots of pages directly to the archive.

Strong personal narratives have enourmous power

I’m a strong advocate of using data and analysis in journalism, but remember that most people don’t relate to numbers in the same way that they relate to visceral personal narratives. We had some stark contrasts in the articles discussed, with STAT’s piece on clinical trials reporting at one end, and the tragic story of the teenager jailed for more than three years while his trial was delayed and delayed at the other. The sweet spot for this type of journalism when you have the data and documents that make your story watertight, but can you tell it through compelling personal narratives.

And remember why we do this work

Investigative journalism should serve the public. Its job is to hold those in positions of power or influence to account, to give voice to the voiceless, and to challenge and correct injustices.