Basic science and statistical understanding

[MrsTerryPratchett]

I'm not a scientist. I do have a BSc. but nothing more impressive. I do wince when I hear some assumptions that people make who don't have basic science (and statistical) knowledge. What basic scientific or statistical principles do you wish the general public (and members thereof) knew about?

I'll start. Causation and correlation. A correlation doesn't prove causation. Particularly when that correlation is a correlation of one. "I had the flu vaccine and got the flu really badly" isn't causation.

2 statistics degrees here, basic probability would be great.

@Mustbe3ormorecharacters

2 statistics degrees here, basic probability would be great.

I've had a quite 'heated debate' with DH (who is an engineer FFS) about the Monty Hall problem. He was wrong.

Sometimes statistics aren't common sense!

Is your husband trolling you?
I am so over discussing this paradox I take the goat.

@MrsTerryPratchett

And that is a COVID thing because people keep saying, "the average age of people who die from COVID is over the average age people generally die" like that's an indication that we shouldn't care. That very much depends on how those numbers are calculated.

Numbers are numbers. The average age of death of those that died from any cause within 28 days of a positive test for coronavirus is 82.4

A raw average hides the spread of data. For example to understand more about the age of death from any cause within 28 days of a positive test of coronavirus we need have some analysis of the ages of death around the mean. One way to do this is to calculate the standard deviation around the mean. Not having the full data it's impossible to do this. However since very few deaths occur in those under 65 and there are very few people who survive longer than 100 it's reasonable to conclude that most deaths will be bunched around the mean of 82.4. Also half of deaths will occur in people over 82.4.

www.mathsisfun.com/algebra/asymptote.html
Further clarification of asymptotic.

@MrsTerryPratchett Good nerd joke, but I'd expect nothing less from someone with your username!

@Mustbe3ormorecharacters

Is your husband trolling you? I am so over discussing this paradox I take the goat.

Nope he was genuinely convinced. It all actually got sorted out at a children's science event where the absolutely lovely children let me beat him repeatedly and convincingly to prove my point.

It was a complete blind spot for him.

I'd expect nothing less from someone with your username!

I’m still seeing anecdote = anecdata.

“My old auntie has COPD”, tested positive but was fine”. Conclusion- we’re being lied to as she should have been really ill based on what they’re telling us.

People really don’t get population data. We see it on here all the time- no that’s not true because I did that and it didn’t happen to me...

That is what I meant @Billie18 and I apologise for my wording.

@MrsTerryPratchett

I'm not a scientist. I do have a BSc. but nothing more impressive. I do wince when I hear some assumptions that people make who don't have basic science (and statistical) knowledge. What basic scientific or statistical principles do you wish the general public (and members thereof) knew about?

I'll start. Causation and correlation. A correlation doesn't prove causation. Particularly when that correlation is a correlation of one. "I had the flu vaccine and got the flu really badly" isn't causation.

Correlation.

Someone tests positive for coronavirus. 28 days they die and we do not have information on the cause of death.

The two events may be correlated but we do not know if the positive test was the cause of death.

That saying that being very overweight makes you more likely to die of Coiv is not fat-shaming, it is an observable fact. Stats are not out to "shame" anyone.

That your nose is part of your respiratory system, and therefore needs to be covered by your mask.

That scientists haven't "changed their mind" on XYZ issue, they've done more research and come to a new conclusion. We used to think smoking was good for you, then in the 1950s and 60s scientists discovered it wasn't.

What PCR is. Everyone seems to be bandying the term about, with no idea what is involved.

I'm a very frustrated biomedical scientist (not practicing) at the moment.

@MrsTerryPratchett

That is what I meant *@Billie18* and I apologise for my wording.

No worries. My wording was hardly perfect

Two major ones

1. The counterfactual - the economy would not be normal in the absence of lockdown as many assume
2. Exponential growth - numbers may be low this week but with a doubling time of x days they will be very high in a matter of weeks so we need to base our current actions on what numbers are projected to do rather than to say there is no need becuase cases are currently low.

Really just baseline critical appraisal, and an understanding of the scientific process.

To be fair to people, probability is a total brain-@#£& - at least I always think so! I'd choose the goat any day (if I could work out the probability of winning it)

That masks stop you spreading covid, not stop you getting it.

Data lags.

The limitations of scientific studies.

The difference between relative and absolute risk.

I think people on this thread might enjoy looking up David Spiegelhalter’s wonderful work on communicating risk.

@PaddyF0dder

Really just baseline critical appraisal, and an understanding of the scientific process.

Do you mean scientific process as in observation, hypothesis, experimentation or something else.

I had a great lecturer at university who used to describe the truth as a statue. The scientists were sculptors who weren't finished. The block of marble was still just a block of marble in certain places and the statue was very true in others. The scientists were chipping away at the untrue marble to reveal the truth. Rarely they would make a mistake and a small piece of truth would end up in the dust on the floor but this didn't mean the pile of marble dust was the truth. And the statue is never finished. But is does become more and more of a likeness of truth.

@HitchikersGuide

To be fair to people, probability is a total brain-@#£& - at least I always think so! I'd choose the goat any day (if I could work out the probability of winning it)

Oh yes, I'd chose goat too! 🐐

Which puts a whole different angle because what it actually shows is teachers are at as high a risk as healthcare and care staff, which, i think we'll all agree is pretty high.

Actually if you look at the stats for healthcare workers they are not at particularly high risk comparatively.

Exponential growth would be my vote.

That masks stop you spreading covid, not stop you getting it.

To be fair, the CDC is now saying that masks can offer some protection based on recent studies.

@bluebluezoo

I’m still seeing anecdote = anecdata.

“My old auntie has COPD”, tested positive but was fine”. Conclusion- we’re being lied to as she should have been really ill based on what they’re telling us.

People really don’t get population data. We see it on here all the time- no that’s not true because I did that and it didn’t happen to me...

This, absolutely. And not just in relation to covid.

@PurpleDaisies

That masks stop you spreading covid, not stop you getting it.

To be fair, the CDC is now saying that masks can offer some protection based on recent studies.

This highlights another one, that advice in a new, rapidly-changing situation is likely to change. That doesn't mean all medical advice is equal. Mask-wearing and advice is different to long-standing advice like smoking. One will change as we know more, the other is really unlikely.

So no, just because 'they' changed their mind on masks doesn't mean 'they' will tell you to stop eating vegetables.

This highlights another one, that advice in a new, rapidly-changing situation is likely to change. That doesn't mean all medical advice is equal. Mask-wearing and advice is different to long-standing advice like smoking. One will change as we know more, the other is really unlikely.

Definitely, and you’ve reminded me of false balance on news reports where a minority view with virtually no evidence to support it is presented as being an equally valid position to take.