Scientists can be intimidating. They can be gruff. They can be enthusiastic. They can talk your arm off.

Interviewing scientists falls into a class of its own.

Most researchers spend their days deeply engaged in a narrow area of inquiry, like protein transport systems inside yeast. Sometimes it's hard to get them to describe the broader importance of their work or to understand why the public needs to know. Some scientists teach courses at universities and do know how to explain their research in an engaging way—and are used to talking for hours without interruption. And with the high level of corporate involvement in science today, many have experience making their work sound fascinating—and lucrative. But often, they're skeptical that reporters can describe their findings accurately or in an appropriate context, and often they've been burned.

Reporters face a variety of challenges: to win scientists' trust, signal what level of information we need, be certain we understand it, and get a sense of both their work's potential and its limitations. Here are a few ways to get the most out of your interviews.

Do Your Footwork

First, if you are reporting on a study, read it. You might be surprised how many reporters just rely on a press release. University public information officers usually have a point of view they want to promote, and sometimes they're wrong. And, if you depend on a summary of a report, you're likely to miss its nuances and its weaknesses.

Even if you're not reporting on a specific study, take the time to read a bit about the scientist you intend to interview and a recent piece of research or two. Your scientist will be impressed—and much more willing to spend time with you.

When you're reading the study, pay extra attention to the introduction and discussion sections. The introduction will put your scientist's research into context. The discussion will describe whatever conclusions the author thinks honestly can be drawn. It should also point out any surprising or odd results, and any problems with the study. Check the data to be sure they match the claims made.

Get a Sense of Your Scientist's Standing in the Field

You can get an idea by looking at the people cited in the footnotes, and also checking out who cited your scientist's work in later publications. Many journals now have online features that make it easy to check citations.

Check for Conflict of Interest

It's a good idea to look for funding sources and affiliations, even if only to see where this particular researcher is coming from. Most journals now list this near the abstract or at the end, just before the footnotes. Also, be alert to competing schools of thought.

Pay attention to:

• Institution(s) involved: How big are they? Do they have a certain take on this area of science?
• Sample size: The bigger the study, the more reliable the results.
• P-value and probability: In order to be statistically significant, p must be < 0.05.
• Outliers—the data that falls on one extreme or another: Do they reveal anything?
• Did a company pay for this study? Donate equipment or materials? Do the scientists own stock or work for a company on staff or as a consultant?
• Who was included: If the study is about humans, did it include only white people? Men? Women? College students? If it was about animals, such as mice, what (if anything) can be inferred about humans?
• Remember: Studies don't prove anything.

Before you do the interview, you'll want to consider:

• How much of an expert is this person? On exactly what?
• Does your scientist tend toward hyperbole? Is she good at metaphor? One way to find this out is to check your scientist’s website and look at things he or she has written. If your scientist is an academic, what type of courses does he or she teach?
• How much do you already know about the subject? Convey this so the scientist comes in at the level you need, instead of offering too much elementary background or too much in the way of technical nuances.
• If you need some of the science basics in your recording, ask for some summary statements and explain why you need them.
• How will you create a relationship of trust? You don't want to agree to let the scientist read your script, but are you willing to go over with that person some of the descriptions and characterizations you plan to use? Just letting them know you have done some background research will often ease scientists' concerns.

Some questions to think about asking:

• What was your most important finding? What surprised you?
• How do you know ___ [whatever claim they are making]? What aspect of your research or other people's work supports it?
• Are your methods generally accepted? Are they unusual or new?
• How do your results compare with others in the area? Do you think they are reproducible or an anomaly? How much consistency is there generally in this area?
• How accurate is your data? What's the level of uncertainty? Were there any areas in which you had to compromise?>/li>
• How sure are you of your conclusions and interpretation? What else could explain your data? Is there anyone who interprets the problem differently?
• Is there controversy in this area? Other schools of thought?
• What got you interested in this area?
• Are there ways you might profit from your ideas, research or results?
• What are the negatives I should know? Who disagrees with your conclusions?
• Do you have pet peeves about the way this area is covered, or about metaphors that are used?
• What's next?

Unless you're a scientist yourself, and sometimes even if you are, you'll need to be sure you understand your scientist correctly. Some ways to check:

• Ask for clarification.
• Restate the concept in your own words.
• Ask "What do you mean?"
• Ask "Can you suggest a good metaphor people will understand?"
• If you think your scientist misspoke, double-check; don't use the recorded statement if it's a mistake. You can say it accurately yourself or go back and re-record.
• Don't try to sound like you understand something when you really don't.

Sally Lehrman is an independent science and health issues reporter for publications ranging from Scientific American to Health magazine, and the science content expert for SoundVision's radio series The DNA Files. She serves as national diversity chair for the Society of Professional Journalists.