-
Asked by to Edward, Ian, Mathew, Naomi, sakshisharda on 19 Jun 2014. This question was also asked by .
-
Naomi Osborne answered on 19 Jun 2014:
It’s not weird at all, cakeak – it’s actually a very good question and sitting for more than 6 hours a day has been linked to a higher risk of not only cancer, but also heart disease and diabetes, but this depends on many other factors as well including exercise. Newspapers do make a habit of scaring people and publicising studies showing only small links between cancers and lifestyle factors, so you do have to take some stories with a pinch of salt!
-
-
Ian Stephenson answered on 19 Jun 2014:
There’s a website (http://kill-or-cure.herokuapp.com) that tracks newspaper stories that report that something causes or reduces cancer . Pretty much EVERYTHING is on the list. Some things are claimed to both cause and cure cancer in different weeks, so obviously something fishy going on (except for smoking – that really does cause cancer!).
Newspapers like a good story, and they have about 6 words of a headline to tell a story. That goes for all kinds of stories:”Mike in love with Claire”, rather than “Mike and Claire have been spending some time together, and while they’re not yet sure if there’s a serious commitment at this stage, they are….” you get the idea.
When newspapers report science stories the often skim over important details. Journalists are rarely scientists!
For example sitting still might increase occurrences of a specific cancer by 1%, but if only 100 people in the word were going to get it anyway, that means everyone sitting still only causes 1 person to get sick. On the other hand sitting still might reduce serious head bashing (which does cause millions of deaths every year) by 50%, in which case its a big win to sit still. However “sitting still causes cancer” is a much better headine that “don’t jump about or you’ll bash your head!”.
Another thing that newspapers love to abuse is is the difference between correlation and causation. What that means is that because two things are linked it doesn’t mean that one causes the other.
For example a particular doctor might have 50% of his patients die. You don’t want to be treated by him right? Well maybe you do if he’s a world expert on a really nasty disease, who only treats patients who would certainly die if he didn’t help them. If you have that disease and you’re probably going to die because of it, then being treated by someone who can give you the best chance of survival is a good choice: 50% is better than 100%.
Correlation is that the two things APEAR to be connected, based on measurements – when one goes up (number of visits to specialist doctor) the other goes up (likelihood of death). People who don’t go to the doctor, and don’t die as often.
However that doesn’t mean doctors CAUSE people to die. CORRELATION (they appear to be connected) doesn’t mean CAUSATION (one makes the other happen). People die because they’re sick, they also go to the doctors because they’re sick.
Often in scientific studies we find a CORRELATION, and it gets published to say “hey this is interesting, someone should figure out whats going on”, but while the scientist would be careful NOT so say one causes the other, journalists aren’t so careful. “Doctor kills patients” is a better story that “doctor gives patients better chance of survival”
It’s a REALLY good exercise to read a newspaper article reporting on something scientific, and actually look at the details. If possible try and find the original scientific work that they’re writing about (journalists are REALLY bad at referencing! they should have a link to the original scientists work). Quite often you’ll read a headline, but if you read the article carefully, you’ll find that the headline is totally misrepresentative of what the researcher actually found out.
Comments
cakeak commented on :
Alright, thank you 🙂 that’s a relief to hear :p