Lies, damn lies, misrepresentation, misinterpretation and STATISTICS

(88 Posts)

A thread for anyone who is infuriated by and/or enjoys reading about this type of stuff. I love it, but annoyingly at the moment I can't think of any examples, other than dull work-related ones. For example, if you're looking at something as a proportion of the whole, then you can't consider one thing in isolation. For example, as Trills mentioned on another thread, let's assume heart disease is the biggest killer of adults (which I believe it is). Let's assume 10% of deaths to adults in 1950 were of heart disease, compared to 42% now. Shocking rise? Probably not.

Plus I will attempt to explain the Monty Hall (think that's the name) problem to anyone who is interested and who doesn't already know it.

Trills Mon 14-Jul-14 22:44:32

Definitely ham.

TheDoctrineOfSnatch Mon 14-Jul-14 22:37:47

"Perhaps the most notorious example of patients being misled about risk occurred in October 1995, when the UK's Committee on Safety of Medicines warned doctors that a new, third-generation oral contraceptive pill doubled the risk of thrombosis. Thousands of women came off the pill, even though the risk had merely increased from a one-in-7,000 chance of getting the disease to a two-in-7,000 chance. The following year saw an additional 13,000 abortions in the UK."

Now that might not be direct cause and effect but is close to what I was trying to say above - reducing one risk might often increase another.

TheDoctrineOfSnatch Mon 14-Jul-14 22:30:49

Thanks!

BoomBoomsCousin Mon 07-Jul-14 02:48:37

Another good BBC article this time on statistics and risk in medicine.

meditrina Tue 27-May-14 09:17:34

Old thread, but I thought that anyone who spots it might also like this BBC story on spurious correlations

Azultrailer Mon 26-Aug-13 01:17:19

I love this thread.

TheDoctrineOfAllan Tue 09-Jul-13 10:27:55

That was an illustrative post but my point was that changing one risk often means changing another, but only one change gets reported.

TheDoctrineOfAllan Tue 09-Jul-13 10:26:09

It should be mandatory for any comment on risk to be balanced by any comments on counter risks.

For example: certain types of SSRIs may increase the risk of foetal heart defects from 2 in 1000 to 4 in 1000 (not exact numbers). However, women of childbearing age who stop taking SSRIs for mild to moderate depression without having alternative treatment arranged suffer an x% increased risk of self harm, worsening depression or whatever.

Debsndan Mon 08-Jul-13 23:31:40

What a great thread. My only contribution is when I was having ivf, I got very interested in the individual success rates of each clinic, including one that actually had a detrimental effect on its patients' fertility. I posted about it on a fertility forum and a woman popped up to say I'd got it all wrong and we all had the same chance of getting pregnant as "it either works or it doesn't, so that's 50/50."
I retreated as my brain had started to hurt a lot and clearly there was no point trying to explain...

rubyanddiamond Mon 08-Jul-13 21:23:38

David Spiegelhalter did an episode of The Life Scientific recently - really interesting if you like this kind of thing:

www.bbc.co.uk/programmes/b02x7h0z

TheDoctrineOfAllan Mon 08-Jul-13 20:36:28

Was it ham, Trills?

Trills Mon 08-Jul-13 20:04:52

87% of women who were given this product for free said it was "quite nice"

Trills Mon 08-Jul-13 20:02:19

You mentioned me in the OP and didn't PM me to tell me this was here!

I had to find out about it on a lottery thread!

Trills Mon 08-Jul-13 20:01:43

Hello.

Aaaaaaaaaaaaaaaaaaaaaaaaaargh! Is it really that hard

"reduced risk" does not = "no risk"
"heightened risk" does not = "certainty"

these are all common words, we use them in the English language all the time. So why do so many people struggle so badly with this concept?

EndoplasmicReticulum Fri 28-Jun-13 22:47:17

I saw a lecture by "Professor Risk" the other day.

understandinguncertainty.org/

I liked the way he looked at things. Especially the idea that if you go for a half-hour run it extends your life by.....half an hour. Great if you like running.

lol thanks for keeping this going!

kim147 Fri 28-Jun-13 21:01:27

The number of injuries went up in 20mph zones year by year because.....they built a lot more 20mph zones.

Now if you looked at the injuries per mile, that would be a very different story.

The number of deaths went up by 70% (!!!) (from 4 to 7)

TheDoctrineOfAllan Fri 28-Jun-13 20:41:56

Um, because they are near schools or other places where more accidents would happen regardless of speed limit?

kim147 Fri 28-Jun-13 20:39:04

More or Less is a great programme.

"20 mph zones are dangerous"

Want to guess the maths behind the headline?

bumping this for any more thoughts or examples

TheDoctrineOfAllan Thu 20-Jun-13 21:22:17

Thanks!

EndoplasmicReticulum Thu 20-Jun-13 20:47:26

just fancied bumping this...

TheDoctrineOfAllan Fri 07-Jun-13 16:22:11

It's like the '8 out of 10 coffee lovers prefer Costa to Starbucks' poster has a definition of coffee lovers somewhere, think it was to do with number of cups per week...

Join the discussion

Join the discussion

Registering is free, easy, and means you can join in the discussion, get discounts, win prizes and lots more.

Register now