Academic research and studies: How they work and why journalists should care
To newcomers the world of academic studies can seem like a foreign land. But as with traveling, one can become quite comfortable over time by learning some of the language and basic customs. The following provides an overview of how studies are produced, their relevance and how they might be used.
What is a scholarly study?
When professors and researchers are not teaching, they are often working on academic studies. These are the basic unit of work in higher education, and are typically a paper of anywhere from several pages up to 50 pages or more. These studies help advance knowledge in a particular field, from AIDS research to the understanding of racial issues in America.
Unlike most journalistic stories or blogs, academic studies are the product of months or years of work; they can include analysis of large sets of data or carefully conducted experiments. A scholar might finish just a few important studies in his or her career, contributing incremental knowledge on a question that may have been studied for decades. Studies aspire to say as much as can definitively be known on a particular question, be it complex or seemingly self-evident. Does money in politics cause corruption? You may consider that an obvious question, but for scholars the answer — not just yes or no, but also why and how — has to be proven with precise weighing of evidence. The essence of the scientific method is to come up with a hypothesis, test it, and then make sure it can be repeated — and that no external factors skewed the results.
Many corporations, commercial research firms, advocacy groups and consulting firms also produce studies and in-depth reports. While these can have news value, bear in mind that the findings of such work are not always independently fact-checked prior to publication, whereas studies produced by academic scholars typically are.
In a world overflowing with information of uncertain quality, it’s hard to find knowledge that is as unbiased, thoughtful and reliable as that contained in the best academic studies. This is why journalists should be familiar with how to read them. Studies can provide a baseline of solid fact where reporting can begin. When journalists call experts to hear their views, having familiarity with the basic research allows for more enlightening conversations and makes stories deeper. Studies almost always suggest a wealth of new angles for journalists to pursue. Further, journalists are connection points between information and the public; it is a journalist’s job to make things clear to the public that are often hidden. Sometimes this means misdeeds by public officials or large corporations. But sometimes important insights can be locked away in research studies and journals. Understanding how to read studies can allow you to bring sunlight to issues and knowledge that might otherwise remain obscure.
Where can I find studies?
Nearly every college and university provides access to research databases such as JSTOR. (Be aware, though, that databases such as JSTOR may not provide access to the latest research.) If you want to look for good studies on the Internet, searching at Google Scholar is another typical route; that can also lead you to the latest research, provided you search by date with the advanced scholar search feature. If you find a study you don’t have access to, you can usually ask your college librarian to help you get access through Academic Search Premier or similar resources. Libraries pay a lot of money for this access — use it. Scholars are also increasingly posting early versions of their studies on their websites. These can typically be downloaded for free, but should be checked against the final versions after publication.
There are an increasing number of “open access” databases that freely offer studies to the public. These include Social Science Research Network (SSRN), Harvard’s DASH repository and the Public Library of Science (PLoS). Such sites are part of the debate about whether studies and the knowledge they contain should remain costly and inaccessible or be freely available. This primer provides an overview of online databases; you should become familiar with many of them. Whatever the outcome of this debate, as a journalist your instinct should be to hunt for the full version of studies and if necessary contact scholars to obtain them. Be aggressive in your pursuit of information. This is the essence of journalism.
How do I read a study?
Studies often have the following structure: abstract; introduction; methods; results; discussion. As tempting as it may be, don’t just read the abstract or summary at the beginning. The more studies you read, the more you’ll get the hang of the language. In fact, the vocabulary and analytical frameworks themselves can become useful for framing stories. Example: Imagine you are reporting on women in the workplace. Is there a “glass ceiling” or a “glass escalator” problem? In the introduction, you’ll generally find that authors review previous scholarship on a particular question. An example is: How does a family’s income relate to a child’s brain development? The literature review tells the reader what has been learned so far and where questions remain.
In the methods and results sections, the researchers describe what data they used, how they analyzed it, and the results. Don’t be intimidated by mathematical jargon; as complicated as it may sound, there’s often a common-sense way to think about it. For example, “regression” basically means trying to figure out how strong the relationship is between the two things — in our example, income and children’s brain development. Ultimately, the researchers are trying to prove not just that the two things are correlated — that they move in parallel — but that there is causation — one thing causes the other.
In the study above, this might mean that a poor child’s development suffers because of lack of parental income. There may be no correlation between the two things or the relationship may be explained by some other factors, such as geography or lack of medical care. The scholar has to sort this all out. Before studies are published in academic journals, they go through a process called “peer review” — other scholars look at the findings to verify that there aren’t any errors. This system ensures that the study is of the highest possible quality, and is the basis on which all good research rests.
While there are an infinite number of questions in the world that can be studied, most research falls into a number of basic types. Below are a few of the categories and loose genres that you should be familiar with:
Data analysis: Social scientists tend to analyze existing datasets, many of them from governmental sources — census data, health information, transportation data and more. This information can be current as well as historical. For example, if you’re interested in education rates among African-American men and women, it’s useful to compare recent and historical levels.
Longitudinal: Long-term datasets can be useful for what are known as longitudinal studies. For example, Social Security data is gathered over long periods, and so can provide insight into the how U.S. residents’ incomes and locations evolve over time. Select groups can also be tracked for years, and thus allow one to understand the relationship between pre-school education and well-being later in life.
Experimental, observational: Studies can also be based on experiments conducted by the scholars themselves; this is often the case in the “hard sciences” such as medicine, chemistry, biology or botany. But they are also used in fields such as psychology and human cognition: For example, medical researchers might want to understand how having Google at our fingertips might affect our memory. The scholars devise and run an experiment — in this case, a memory test of a group of study subjects — and then analyze the results. When studying medical questions (say, the impact of a new HIV/AIDS treatment) it’s essential to have a what’s known as a control group — subjects who are given a placebo rather than the substance under study. If the control group is under the control of the researchers, the study is known as experimental; if they’re not, it’s observational. Social scientists are also running more controlled trials in the field: For example, how hearing the Spanish language in America affects perceptions and political attitudes.
Metastudies: These are, in essence, studies of studies. These can be helpful when there is a lot of research on a particular topic — for example, how studying at a diverse college might affect civic engagement later in life. Using meta-analysis, the researchers take the results of the previous studies and combine them, looking for patterns and, in essence, finding “the truth” on a particular question.
Survey-based research reports: While not technically studies, surveys frequently contain valuable information for journalists. They’re often conducted by organizations with a longstanding expertise in the area such as the Pew Research Center, a gold standard for this sort of study. Questions can range from the American public’s changing views of government to the rise of women in higher education.
Reports: Governmental and nonprofit research organizations often produce reports that synthesize a wide variety of data and look at particular questions — for example, the importance of infrastructure investment, the disparities faced by minorities in California, or trends in college spending. Reports can be supported by an organization with a particular mission or point of view, but that doesn’t necessarily invalidate their findings, particularly if they’re published by a credible organization.
Papers: While there’s no precise definition, papers tend to be written by a smaller number of researchers — often just one or two — on broader societal questions. Examples include a rethinking of economic development in poor countries or the case for banning subprime mortgages. Papers are often produced by think tanks such as the Brookings Institution or the RAND Corporation, but can also come out of academia.
The reading of scholarly studies can be important to get context — to know what smart people have thought about a particular issue. But they can also be cited and summarized in your story or blog. You might strongly consider sending the author of a study an email inquiry or calling him or her to be sure you correctly understand the findings of the study. No matter what, be sure to give the scholar full credit by mentioning the institutions the study came from and the journal in which it was published. And link to the study so your audience can go further into the question. It will improve the depth of your journalism and demonstrate transparency and respect for your audience.
You might first think about how to localize a study. Here are 10 examples of such study-based ideas. Furthermore, the following are examples of high-level stories that use research in a variety of different ways. You’ll see that study findings can be used for context or background; for the purpose of providing alternative perspective on issues; or as hooks for stories themselves.
- “Wasting Time Is New Divide in Digital Era,” The New York Times
- “The Killing Agency: Wildlife Services’ Brutal Methods Leave a Trail of Animal Death,” Sacramento Bee
- “The Unpersuaded: Who Listens to a President?” The New Yorker
- “Whole Food Blues: Why Organic Agriculture May Not Be So Sustainable,” Time
Studies can also be used to provide crucial context and fact-check claims.
Take, for example, the 2012 story about a congressman’s explosive comments casting doubt on the possibility of pregnancy from rape (he was asked about abortions in such situations.) An initial online article such as this one at NPR includes reaction quotes and reporting on the outrage stirred up. But a reporter could also, in theory, quickly search the PubMed database and locate this peer-reviewed study giving the facts behind the issue. “Rape-related pregnancy occurs with significant frequency,” the study notes, putting the figure at 5% and estimating more than 32,000 rape-related pregnancies across the population annually. This Pacific Standard blog post — written shortly after the story broke — performed that research. The data highlighted in the study can then be used to anchor all subsequent reports on the controversy, improving public understanding and not allowing the discussion to revolve around simple “he said, she said” exchanges between partisan sources.
What are the hazards of using studies?
The greatest danger of using studies in journalism is through oversimplifying the findings or misinterpreting them altogether. Researchers often present their findings with many, many reservations — they want to get things right, and so if they have any doubts or see areas for future research, they’ll mention them. You might wish that they’d made a direct, unqualified assertion that fits in a headline, but it’s often not that simple. As a journalist, you may find that there is no way to accurately state the study’s conclusions in a few sentences. In those cases, it is essential that you tell your audience that the study’s findings are limited by certain factors.
By all means, call or email the author and ask questions until you are satisfied that you understand the study’s findings and implications. Researchers are often very helpful to media members, as they have every interest in seeing their work presented accurately to the public.
Keep in mind that scholars consider research to be a process that is constantly unfolding. While some questions can be definitively answered — for example, that certain drugs can slow the spread of HIV — on almost all subjects there is always more research to be done.
As a journalist, it is crucial to avoid what some call the “single-study syndrome” — basing a story entirely on the results of one study, without exploring alternative research angles. While the peer-review process should ensure that only the best and most reliable research is published, some studies can be “outliers” and not necessarily represent general facts about an issue. This is most often a problem with studies based on experiments, because their precise design can greatly influence the results obtained. So if you see an experiment-based study that has produced unusual results, proceed with caution.
Finally, there are always questions about potential bias, particularly as a consequence of who funded the research. This can be difficult for non-experts to assess. Typically, scholars will disclose any conflicts of interest at the bottom of the study. You should be especially wary of research that does not appear in a peer-reviewed journal, or research that is funded by commercial firms that might have a vested interest in the results.
What should a journalist do after reading the study?
You should make every effort to get in touch with the author. But that may not be the end of your follow-up. One thing you can do to assess the validity of the findings independently is to go to the study’s citations — all of the footnotes contained in the bibliography at the bottom — and contact the authors of the other studies that are cited. You might email them and say, “You are cited in this study. What do you think of the findings?” If you are going to be reporting in this area over time, you might also begin signing up for email alerts for the journals that are related to this field. That way you can stay up to date with what the academic community is saying about aspects of your beat.