Author Topic: Statistics and bacon  (Read 5290 times)

bluehillside Retd.

  • Hero Member
  • *****
  • Posts: 19724
Re: Statistics and bacon
« Reply #25 on: April 17, 2019, 04:45:51 PM »
NS,

Quote
So they are a perpetual virgin?

Given all that bacon they're eating, probably...
"Don't make me come down there."

God

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #26 on: April 17, 2019, 04:46:03 PM »
In the case of science reporting, your assumption just makes you assuming, and based on evidence against it.
You have to set the bar somewhere - you cannot assume that no-one is able to understand basis English and the meaning of words put together into a simple phrase like increases the risk.

And you are correct, you have to make an assumption either way - you either assume that people are able to understand simple English or you assume that people aren't able to understand simple English - I think the former is better as it doesn't stilt discussion and it means you don't constantly patronise people by assuming they can't understand basic English.

How are you getting on with that maths working out?

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #27 on: April 17, 2019, 04:48:25 PM »
You have to set the bar somewhere - you cannot assume that no-one is able to understand basis English and the meaning of words put together into a simple phrase like increases the risk.

And you are correct, you have to make an assumption either way - you either assume that people are able to understand simple English or you assume that people aren't able to understand simple English - I think the former is better as it doesn't stilt discussion and it means you don't constantly patronise people by assuming they can't understand basic English.

How are you getting on with that maths working out?

You can make that assumption. Reading reporting of science in the media might make you wonder but on you go with you thinking it's all good with your assumption, and licking the boots of science reporting in the media.

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #28 on: April 17, 2019, 04:49:59 PM »
NS,

Given all that bacon they're eating, probably...
I think that's a bit rasher!

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #29 on: April 17, 2019, 04:50:45 PM »
Prof,

But look at the ambiguities in that "every"! Clearly it can't actually mean "every" or just one slice eaten by an otherwise lifelong vegetarian would have the same effect wouldn't it? So does it mean the 20% increase is for people who eat just one rasher every day? And if it does, does that mean that eating more than one rasher every day increases the risk even more? I think we should be told!
Yes - that's non-sense reporting. A single slice of bacon doesn't increase the risk of bowel cancer by 20% (or at least there is no evidence in the study to support that notion). What the study found was that people who eat red and processed meat four or more times a week had a 20% greater likelihood of being diagnosed with bowel cancer compared to those who eat read and processed meat less than twice per week.

The Independent reporting (although perhaps not the headline) is fairly clear:

https://www.independent.co.uk/news/health/bowel-cancer-bacon-sausage-red-meat-processed-oxford-health-a8872716.html

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #30 on: April 17, 2019, 04:52:26 PM »
You can make that assumption. Reading reporting of science in the media might make you wonder but on you go with you thinking it's all good with your assumption, and licking the boots of science reporting in the media.
Not only do you seem unable to understand simple English, you have now ceased to be able to write it either. Have you morphed into Vlad ;)

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #31 on: April 17, 2019, 04:56:07 PM »
Not only do you seem unable to understand simple English, you have now ceased to be able to write it either. Have you morphed into Vlad ;)
One of your usual tropes when you have lost the argument.

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #32 on: April 17, 2019, 04:57:12 PM »
One of your usual tropes when you have lost the argument.
What argument?!?

How's that maths coming along? Ready to share?

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #33 on: April 17, 2019, 05:00:06 PM »
What argument?!?

How's that maths coming along? Ready to share?
The one you lost when you admitted to making assumptions about science reporting.

And btw I think you mean arithmetic.

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #34 on: April 17, 2019, 05:05:46 PM »
The one you lost when you admitted to making assumptions about science reporting.
I didn't make any assumptions about science reporting, I made an assumption about people's ability to read simple English - in other words to understand what the phrase increases the risk means.

That is perfectly reasonable I'm afraid as to assume that people cannot understand what the phrase increases the risk (or any other simple phrase) means would mean we'd spend our who time constantly explaining simple phrases to each other, which would kill any kind of discussion and would be deeply patronising.

I make no apology for assuming that posters here are able to understand what increases the risk means. Wiggs OP wasn't about that, but about what the stats around it meant.

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #35 on: April 17, 2019, 05:08:55 PM »
And btw I think you mean arithmetic.
Into pedantry now - sorry I must not assume you know what pedantry means, so here is the dictionary definition:

'excessive concern with minor details and rules'

Oops better explain what 'excessive', 'concern', 'minor details' and 'rules' means now - see where we get when we don't assume people are capable of understanding simple English NS.

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #36 on: April 17, 2019, 05:09:49 PM »
I didn't make any assumptions about science reporting, I made an assumption about people's ability to read simple English - in other words to understand what the phrase increases the risk means.

That is perfectly reasonable I'm afraid as to assume that people cannot understand what the phrase increases the risk (or any other simple phrase) means would mean we'd spend our who time constantly explaining simple phrases to each other, which would kill any kind of discussion and would be deeply patronising.

I make no apology for assuming that posters here are able to understand what increases the risk means. Wiggs OP wasn't about that, but about what the stats around it meant.
I really hope you never pass a mirror, asumptionist.

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #37 on: April 17, 2019, 05:10:58 PM »
Into pedantry now - sorry I must not assume you know what pedantry means, so here is the dictionary definition:

'excessive concern with minor details and rules'

Oops better explain what 'excessive', 'concern', 'minor details' and 'rules' means now - see where we get when we don't assume people are capable of understanding simple English NS.
Long winded but a perfect illustration of the word irony from you

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #38 on: April 17, 2019, 05:11:43 PM »
And btw I think you mean arithmetic.
Would you like to share your arithmetic workings with us please - the one that construed a 20% increased risk on a base risk of 6% as: "an increase of .2% v the 94%"


ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #39 on: April 17, 2019, 05:13:15 PM »
I really hope you never pass a mirror, asumptionist.
Blimey spelling is suddenly the next element after inability to understand or to write simple English.

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #40 on: April 17, 2019, 05:17:43 PM »
NS,

Well, maybe he’d be in good company if he does:


Catholic Mariology

"The four dogmas of perpetual virginity, Mother of God, Immaculate Conception and Assumption form the basis of Mariology."

https://en.wikipedia.org/wiki/Catholic_Mariology
Not sure I think that's such good company.

bluehillside Retd.

  • Hero Member
  • *****
  • Posts: 19724
Re: Statistics and bacon
« Reply #41 on: April 17, 2019, 05:18:15 PM »
NS,

Quote
I think that's a bit rasher!

Oh rind yer neck in why doncha... ;)
"Don't make me come down there."

God

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #42 on: April 17, 2019, 05:20:20 PM »
Where the thing that is being discussed (e.g. eating red and processed meat every other day) is probably very common and likely the majority situation I think it might be better, and more positive, to turn the reporting on its head. In other words to say that if you reduced consumption you would reduce your risk of cancer.

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #43 on: April 17, 2019, 05:26:14 PM »
Blimey spelling is suddenly the next element after inability to understand or to write simple English.
woo it's a spelling thing to make an argument. How exciteing!

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #44 on: April 17, 2019, 05:27:15 PM »
NS,

Oh rind yer neck in why doncha... ;)

We are on a streaky here
« Last Edit: April 17, 2019, 05:34:32 PM by Nearly Sane »

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #45 on: April 17, 2019, 05:30:59 PM »
Where the thing that is being discussed (e.g. eating red and processed meat every other day) is probably very common and likely the majority situation I think it might be better, and more positive, to turn the reporting on its head. In other words to say that if you reduced consumption you would reduce your risk of cancer.
said a sponsor for the eating of pigs industry


bluehillside Retd.

  • Hero Member
  • *****
  • Posts: 19724
Re: Statistics and bacon
« Reply #46 on: April 17, 2019, 05:31:26 PM »
Prof,

Quote
Yes - that's non-sense reporting. A single slice of bacon doesn't increase the risk of bowel cancer by 20% (or at least there is no evidence in the study to support that notion). What the study found was that people who eat red and processed meat four or more times a week had a 20% greater likelihood of being diagnosed with bowel cancer compared to those who eat read and processed meat less than twice per week.

The Independent reporting (although perhaps not the headline) is fairly clear:

https://www.independent.co.uk/news/health/bowel-cancer-bacon-sausage-red-meat-processed-oxford-health-a8872716.html

Yes I get that - drill down a bit and you can get the sense of it. You have to admit though that linking "every slice" with a 20% risk increase is potentially at least very misleading.

Reminds me a bit of the headlines a while back that cats falling from above six-storey heights suffered fewer injuries than those falling from lower heights. Various explanations were offered too - maybe the longer fall gave them more time to reorientate feet down, maybe they reached a velocity at which their fur would have a parachute effect etc. Turned out they were more likely to die than to be injured, and dead cats don't get taken the vet so there were fewer reports of injures. Funny things sometimes, statistics.       
« Last Edit: April 17, 2019, 05:41:41 PM by bluehillside Retd. »
"Don't make me come down there."

God

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #47 on: April 17, 2019, 05:34:55 PM »
said a sponsor for the eating of pigs industry
Last time I looked not all red and processed meat was produced from pigs.

ProfessorDavey

  • Hero Member
  • *****
  • Posts: 18013
Re: Statistics and bacon
« Reply #48 on: April 17, 2019, 05:37:11 PM »
Prof,

Yes I get that - drill down a bit and you can get the sense of it. You have to admit though that linking "every slice" with a 20% risk increase is potentially at least very misleading.
Yup completely misleading, particularly as the comparison was between people who eat on average 76g of red and processed meat a day compared to people eating 21g a day.

You also have to be careful that the study fully factors out other linked effects - e.g. obesity. It could be that being overweight is the driving factor rather than red and processed meat eating per se, and that people who tend to eat the most red and processed meat tend also to be the more overweight.

You need to be very careful that association and causation aren't confused.
« Last Edit: April 17, 2019, 05:40:44 PM by ProfessorDavey »

Nearly Sane

  • Administrator
  • Hero Member
  • *****
  • Posts: 65855
Re: Statistics and bacon
« Reply #49 on: April 17, 2019, 05:44:46 PM »
Last time I looked not all red and processed meat was produced from pigs.
Feck me, a humour bypass has been successful