Communicating their work effectively, and in a way that’s understandable to the general public, is something that scientists and even professionals in communicating science seem to struggle with. The balance between making a headline jump out at readers and grab their attention, yet convey the truth accurately and get into the real detail of a topic can be difficult with scientific news, as no one wants to hear about all the “maybes” and “potentials”, or the nitty gritty and often unexciting scientific methods. We want to hear about exciting new discoveries. A large part of the communication issue is due to a sort of “language barrier” between the academic scientific world, and the rest of us. There are whole lists of words that a scientist might use without even thinking about, under the impression that they’re easily understandable, having used the word in their work for years, that actually a lot of us might not grasp their definition of. Likewise, there are a huge number of phrases used by scientists to mean one thing, but mean something entirely different to a non-scientist. Even between the different areas of science, communication can be a problem, as physicists can use a word to mean one thing, while a biochemist might mean something completely different. So, I’ve decided to write a series of blog posts about some of the phrases and words you might see in the news or hear scientists saying, that actually don’t mean what you think they do. Equally, this might help scientists and science communicators to realise how some of the words they use don’t actually mean to the public what they’re trying to convey.
First, I’m going to talk about the use of the word “link” or “connection”:
“Scientists find genetic link to obesity”
“Paracetemol linked to heart attacks”
“Genetic link to homosexuality found”
These are only a few examples of headlines you might have seen recently. If you don’t read the rest of the article (and even sometimes if you do), you can come away thinking that obesity and homosexuality is completely and unquestionably genetic, that paracetamol causes heart attacks, and that these are certain, proven facts – the scientists say so, after all. And I wouldn’t blame you for thinking that. In actual fact, what these headlines really mean is slightly different.
I’ll start by talking about the mysterious “genetic link”. While I’ll admit that the use of the word “link” doesn’t actually tell us that genetics are the cause of a certain condition or trait, it doesn’t do well to persuade us otherwise either. A more suitable headline might be “Scientists find that some people are gay, and also have a certain gene, although others don’t and we’re not entirely sure if it’s just a coincidence or if one is caused by the other”, but I guess that wouldn’t be very catchy. What this usually means is that scientists have conducted a study, in which they’ve studied a sample group of (hopefully a large number of) people, some of whom are gay, and some of whom are not. The scientists then examine the genomes (the entirety of the DNA) of these people and try and see if they have anything in common. If there’s a certain sequence of DNA that appears more often in people who are gay than in those who aren’t, then they might call it a “genetic link” or a “potential gay gene”. While statistical analysis is used to work out the probability of the results being purely due to coincidence, a lot of the time we’re still not 100% certain the gene is the cause of this characteristic.
For example, the variation that we find correlates with a certain trait might not be the one causing the trait. We might pick up on one change in a person’s genome, and see that it coincides with a certain trait, but it could actually be another change to their genes, somewhere nearby, that causes it, and because this second change is so close to the first, they tend to occur at the same time, are inherited together, so appear in the same individual. So while this gives us an idea that people with this sequence are statistically slightly more likely to be gay than those without it, it doesn’t tell us that the sequence is the cause of their sexuality (there may be a correlation between the two, but no clear causative link).
When it comes to links between certain behaviours and illnesses, it’s a similar matter, although this isn’t to say that there isn’t usually a firm basis for these kind of claims. One example is the link between smoking and lung cancer: in this case, the incidence of lung cancer amongst smokers is much higher than in the general population; we know that smoking does increase your chances, and we also know some information about the mechanism by which this occurs. While it doesn’t mean that everyone who smokes will develop lung cancer, the link and even the causation is pretty clear and very strong. With some newer claims however, the information is a little less clear. We’re always seeing headlines telling us that our favourite food is linked to cancer, or that hayfever medication can cause alzheimers, but we need to be a little careful about how heavily we take these claims.
Sometimes, when a large number of people who participate in a certain behaviour or have certain personality traits do get a certain illness, it’s not actually this exact trait which causes the illness. It may be that people who participate in this behaviour are also likely to participate in a different behaviour, or lead a certain lifestyle and it’s that behaviour which increases their risk of the disease. Or maybe, it’s just a coincidence.
To illustrate this, I’ll give you a few weird correlations:
Now, both of these correlations are true according to the people surveyed. But this doesn’t mean there’s any kind of connection between whether or not you like Neil Patrick Harris and whether or not you pee in the shower (or is there?), and similarly, not all genetic links tell you something about the cause of a certain trait or condition. A link is a link. Not a certainty, not necessarily a cause, but a potential clue to the cause of a certain characteristic, which with a little more work, maybe will be proven, and tell us more about our genetics, or the effects of certain medications or behaviours.