Take Bearings (Navigation-based Information Evaluation)

Reading the Room on the Mask Issue

Suppose you encounter an article questioning whether N95 masks are effective. Before examining its detailed argument, what background do you need to assess it?

If you’re already well grounded in what the current research says on an issue and adept at computing the Brownian motion formula, fine, go ahead, dive in. However, if you’re not, the first thing you should do is take bearings, a concept drawn from navigation. A lot of what happens when we search is like parachuting into unfamiliar territory. When you don’t know the terrain, it’s easy to get spun around in circles. Only foolhardy paratroopers go traipsing off toward the first mountain peak they see. Instead, they use their compass to “take bearings”—measuring the angle between true north and their desired destination, so that even with poor visibility, they can look at their compass and proceed on a bearing of, say, forty-five degrees and not get lost when fog sets in. Obviously, taking bearings on the web is not an exact science, and certainly not as exact as measuring an angle. But the premise is analogous—when landing in unfamiliar terrain, head off only after you’ve gotten the lay of the land.

On a topic where you lack vital background information, how should you start? Let’s return to the issue of masks to find out.

To get a bird’s-eye view of the terrain, we might forgo papers that focus on a single topic and see what a group like the Occupational Safety and Health Administration (OSHA) has to say. Take this page, which gives a clear summary of the factors in play:

Will an N95 respirator protect the wearer from the virus that causes COVID-19?

Yes, an N95 respirator is effective in protecting workers from the virus that causes COVID-19. . . . The National Institute for Occupational Safety and Health (NIOSH) tests respirators using particles that simulate a 0.3 micron diameter because this size particle is most likely to pass through the filter. If worn correctly, the N95 respirator will filter out at least 95% of particles this size. An N95 respirator is more effective at filtering particles that are smaller or larger than 0.3 microns in size.

Who is OSHA, you wonder, and what the heck does it know? Well, if you practice the I in SIFT, you’ll find OSHA is the government entity responsible for setting workplace safety standards—for things like masks. (It started doing this long before COVID-19, often in workplaces that have contact with hazardous substances.) And to do that, it relies on the advice of not just one expert but the larger expert community. So you’re getting the benefit of OSHA summarizing the expert landscape for you.

But maybe you don’t trust OSHA. Fine. Who do you trust, then, to accurately summarize not just a single researcher’s opinion but the state of knowledge in the field? You should be careful where you invest your trust, but you have to invest it somewhere. If you throw up your hands and say, “Maybe the truth is in the middle, or maybe it’s all lies,” then you’re exactly where the media manipulators want you to be.

Maybe in this case you go and see what the American Medical Association (AMA) says. This is a professional association of doctors in the United States. The value of its guidance depends on how well it represents the evidence and uncertainty in the field. In the guidance discussed here, it, too, said that N95 masks work, while recommending priority for health-care workers at that time.

If you don’t trust the AMA, pick a publication with a dedicated health beat (e.g., one that has a reporter with extensive experience reporting on health issues). Again, you’re looking for an article where a reporter relies on a variety of respected experts in the field and tries to summarize the current state of knowledge regarding the issue of certainty (how certain experts are that masks work) and the issue of consensus (how much agreement there is on this issue).

Maybe you think the US institutional infrastructure is hopelessly compromised by Democratic rule (it’s not) or Republican rule (it’s not). But if you do think that, you’re free to look at the advice issued by other countries. Does that advice differ substantially from the United States? To check, you may want to look at the Lancet, a medical journal based in the United Kingdom.

We hope you get the point. You can choose any one of these, depending on what strikes you as the most trustworthy. But any one of these, as a summary of the current state of knowledge in an entire field, is going to be a better starting point than a single paper. That single paper may be useful once you’re ready to do a deep dive. But when you first start, you need a bigger, more inclusive map of the terrain, and you’re more likely to get that from a broad research summary. Start digging deep into the weeds of comparing individual papers in an area outside your expertise, and you’ll quickly find yourself at the bottom of the trust compression sinkhole.

Note that even here we can see the importance of consensus. OSHA is saying the same thing as the AMA, which is saying the same thing as most experts. If OSHA had said something very different from the AMA, or the AMA had disagreed with experts quoted in health reporting, we might recommend slowing down a bit. Instead, we found strong consensus across multiple sources of expertise. Throw that bottle, along with the note it contains, back into the lake.