Data for Journalists

A Practical Guide for Computer-Assisted Reporting

Brant Houston

• Editore: Routledge
• Anno: 2018
• ISBN:
  9781351249294
• Edizione: 5
• Argomento: Sprachen & Linguistik
• Categoria: Journalismus

This straightforward and effective how-to guide provides the basics for any reporter or journalism student beginning to use data for news stories. It has step-by-step instructions on how to do basic data analysis in journalism while addressing why these digital tools should be an integral part of reporting in the 21st century. In an ideal core text for courses on data-driven journalism or computer-assisted reporting, Houston emphasizes that journalists are accountable for the accuracy and relevance of the data they acquire and share.

With a refreshed design, this updated new edition includes expanded coverage on social media, scraping data from the web, and text-mining, and provides journalists with the tips and tools they need for working with data.

1

What Data Journalism and Computer-Assisted Reporting Is and Why Journalists Use It

Journalists need to be data-savvy. It used to be that you would get stories by chatting to people in bars, and it still might be that you’ll do it that way some times. But now it’s also going to be about poring over data and equipping yourself with the tools to analyze it and picking out what’s interesting. And keeping it in perspective, helping people out by really seeing where it all fits together, and what’s going on in the country.

—Tim Berners-Lee, creator of the World Wide Web (2010)

The open question in 2014 is not whether data, computers, and algorithms can be used by journalists in the public interest, but rather how, when, where, why, and by whom. Today, journalists can treat all of that data as a source, interrogating it for answers as they would a human.

—Alexander Benjamin Howard, The Art and Science of Data-Driven Journalism (2014)

While computer-assisted reporting (CAR) has been much more widely practiced since the 1980s, it is since the beginning of the 21st century that the critical importance of journalists having the ability to analyze and visualize data has been globally recognized.

Fortunately, during the same time, it has become much easier to do. Many software tools have become simpler to use. A rich variety of data is now online and often very simple to download. Storage space is immense on hard drives, flash drives, and in the cloud. The computing power on a laptop, tablet, or mobile phone dwarfs the power available only a few years ago.

The ways to visualize data for better understanding and analysis are numerous and increasing. Furthermore, a wave of computer programmers/coders has joined with journalists to tackle the problems of capturing data from the web, cleaning and organizing it, and creating fascinating interactive presentations to be shared with the public and that encourage citizen participation and analysis.

However, many fundamental truths remain. Databases are created by people—or by programs created by people. Thus, databases have errors and omissions that people have made and those flaws must be noted and corrected. Every database also is a slice in time and thus, potentially outdated the moment it is acquired and used.

Also remember that a database alone is not journalism. It is a field of information that needs to be harvested carefully with insight and caution. It needs to be compared and augmented with observation and interviews.

It is still important to determine the accuracy of a database before using it. Equally important is the careful analysis of a database since one small error can result in monstrously wrong conclusions. The idea of uploading data on the web and hoping the public or volunteers will consistently make sense of it with reliable analysis has proven unreliable. In fact, journalists—not advocates—are needed more than ever to deliver a well-researched understanding of information and data, and to write and visualize a compelling story.

The Fundamentals Remain

Despite changes in technology and the availability of mega-data, some scenarios have not changed.

For example, as a local reporter in the United States, you may want to look into how many inmates are in jails because they do not have the money for bail, which means they are in jail until trial. You see that a recent audit of jails points out that many persons are in jail not because they have been convicted, but because they simply do not have the finances to pay the bail set for them. Furthermore, it appears anecdotally that judges are setting higher bails for black and Hispanic males than white males.

With a little research, you find that the county jails keep records on inmates that include the amount of bail in each case, the charges against the inmate, and each inmate’s race and gender. After a series of meetings, the officials will agree to give you the database with the information you need. Irritatingly, they do not want to put it online for you to retrieve, but they will give you a DVD containing the database.

By the next day, you do your first analysis of the data and see that the bail for black and Hispanics males is often double that for white males, even when they are charged with the same crime and have the same criminal and personal background. Over the next few weeks, you check through the records and gather more details. You recheck your information, look at other documents, conduct interviews, and write the story. The work culminates in a significant article that presents a systematic look at justice gone wrong. The best weak answer the officials have is the system discriminates against the poor, not just blacks and Hispanics.

Or consider another scenario: you want to know how weak security is at your nearby metropolitan airport. So you get local police reports or download recent information from the Transportation Security Administration about dangerous materials seized as passengers go through security. You begin analyzing the database, which consists of counting the number of violations at local airports in recent years and then examining closely the details of those violations.

You quickly find serious and surprising violations in which guns, knives, and other weapons are seized. You follow up with research on the web and interviews with airport officials, law enforcement, and airline companies. You review reports by government investigators posted on the web. Within days, you have an important story that the public needs to know.

In fact, more than 100 news organizations used local police databases while doing airport security stories in the weeks, months, and years following the terrorists’ attacks in the United States on September 11, 2001. More recently, journalism students at Medill wrote this very story and created an online database to go with it.

On a more immense and global scale, consider this example:

A newsroom receives a leak of a millions of documents revealing the use of offshore bank accounts by the wealthy or international criminals.

Rather than simply post the documents to an online site, the newsroom carefully reviews the authenticity of the data, converts it from unstructured data (text) to structured data that can be analyzed and visualized, and then shares it with hundreds of journalists across the world to report and verify on the data before publishing a word.

This has actually happened several times in the past decade at the International Consortium of Investigative Journalists, a non-profit newsroom in Washington D.C.

At the base of these techniques is a practice known as CAR—also called data journalism—which is often the place to begin if you are a journalist without computer science background. As you will learn in this book, CAR has become a part of everyday journalism. Journalists use these and other techniques for daily reporting, beat reporting, and for the large projects that win national and international awards. In the last two decades, journalism awards have gone to the International Consortium for Investigative Journalism for its investigations using big data leaks; The Washington Post for stories on police shootings and child abuse; and to the small newspaper, the Bristol (Virginia) Herald Courier, for the work of reporter Daniel Gilbert who built and used a database as part of his investigation into natural-gas royalties owed to thousands of landowners in southwest Virginia.

CAR does not refer to journalists sitting at a keyboard writing stories or surfing the web and social media. It refers to downloading or building databases and doing data analysis that can provide context and depth to daily stories. It refers to techniques of producing tips that launch more complex stories from a broader perspective and with a better understanding of the issues. A journalist beginning a story with the knowledge of the patterns gleaned from 150,000 court records is way ahead of a reporter who sees only a handful of court cases each week.

The techniques of CAR do not replace proven journalistic practices. It has become a part of them. It also requires greater responsibility and vigilance. The old standard—“verify, verify, verify”—that one learns in basic reporting classes becomes more critical. “Healthy skepticism” becomes ever more important. The idea of interviewing multiple sources and cross-referencing them becomes ever more crucial.

“Computers don’t make a bad reporter into a good reporter. What they do is make a good reporter better,” Elliot Jaspin, one of the pioneers in CAR, said three decades ago.

Many practicing journalists have sought training in the past three decades and become proficient in the basic skills of CAR. They have overcome computer and math phobia, and they now put these skills to use on a daily basis, leading to more precision and sophistication in their reporting.

To quote Philip Meyer, another pioneer in database analysis for news stories, “They are raising the ante on what it takes to be journalist.”

Aiding in the progress and acceptance of these skills has been the proliferation of the web and social media, the development of inexpensive and easy-to-use computers and software, the increased attention to the value of data and techniques of data and visual analysis in newsrooms, and the use of algorithms in finding patterns and creating content.

CAR and the other approaches to data—such as precision journalism or computational journalism—are no longer a sidebar to mainstream journalism. They are essential to surviving as a journalist in the 21st century. The digital tools won’t replace a good journalist’s imagination, the ability to conduct revealing interviews, or the talent to develop human sources. But a journalist who knows how to use analytical and visual software in day-to-day and long-term work will gather and analyze information more quickly, and develop and deliver a deeper understanding. The journalist will be better prepared for interviews and be able to write with more authority. That journalist also will see potential stories that would have never occurred to him or her.

That journalist also achieves parity with politicians, bureaucrats, advocates, and business people who have enjoyed many advantages over the journalists simply because they had the money and knowledge to utilize databases and digital information before journalists did. Government officials and workers have long been comfortable entering information into computers and then retrieving and analyzing it. Businesses, small and large, routinely use spreadsheet and database software. Advocacy groups frequently employ databases to push their agendas.

Without a rudimentary knowledge of the advantages and disadvantages of data analysis, it is difficult for the contemporary journalist to fully understand and report on how the world now works. And it is far more difficult for a journalist to do meaningful public service journalism or to perform the necessary watchdog role.

As long ago as 1990, Frank Daniels III, former executive editor of The Raleigh News & Observer, recognized the challenge. He began his newspaper’s early and oft-lauded push into CAR because the 1990 campaign of then Senator Jesse Helms was profoundly more computer-sophisticated than Daniels’s own newspaper. “It made me realize how stupid we were, and I don’t like feeling stupid,” Daniels recalled.

Daniels was right about the bad position in which journalists had put themselves. For years, journalists were like animals in a zoo, waiting to be fed pellets of information by the keepers who are happy for journalists to stay in their Luddite cages. But a good journalist always wants to see original information because every time someone else selects or sorts that information, they can add spin or bias, which sometimes can’t be detected. CAR can help prevent that from happening.

Many journalists and journalism students now learn the basic tools of CAR because they realize that it is the best way to get to the information because most governmental and commercial records are now stored electronically. Despite security concerns and efforts to keep information secret, there still are a mind-boggling and growing number of databases on U.S. and international websites. So without the ability to deal with electronic data, a journalist is cut off from some of the best, untainted information. The old-fashioned journalist will never get to the information on time—or worse, will be brutally trampled by the competing data savvy media.

For a journalist or journalism student, this knowledge also is crucial in the competition to getting a good job. At many news organizations, an applicant that has these skills—which are far more than the ability to search the web or use social media—gets his or her resume moved to the top of the list.

A journalist does not have to be a programmer or someone who knows software code, although that also can make a huge difference. But a journalist who can use a spreadsheet or database manager is free to thoroughly explore information, reexamine it, and reconsider what it means in relation to interviews and observations in the field. The journalist can take the “spin” off the information and get closer to the truth. A journalist may not be a statistician, but a good journalist knows enough about statistics to know how easy it is to manipulate them or lie with them. In the same way, if a journalist understands how data can be manipulated, he or she can better judge a bureaucrat’s spin on the facts or a government’s misuse of a database.

Journalists have found, to...