Covid 19 outbreak - arguing about Stats (again)

[MikeF]

I fear that the standard response to asking for a source or for evidence is "Look it up yourself."

Except I have and given the source.

[Oldjohnw]

I fear that the standard response to asking for a source or for evidence is "Look it up yourself."

Except I have and given the source.

You have. But you were asking someone else for a source. I was not particularly optimistic about the likelihood of your getting such.

[Ben@Forest]

According to the Telegraph (quoting an NHS update) there is not currently a higher occupation of intensive care beds than would be normal at this time of year. And apparently even in April there was no more than 80% occupancy. At the weekend Whitty and Vallance warned that beds would be full by November 20th.

Of course it's an imponderable. If beds aren't full by November 20th it will be regarded as a measure of how successful lockdown is. And we'll never really know.

[Oldjohnw]

Covid
Brexit
Trump
And now the terror level is raised. And I can't even head to the hills.

[Jdsk]

According to the Telegraph (quoting an NHS update) there is not currently a higher occupation of intensive care beds than would be normal at this time of year.

TTBOMK that dataset is currently suspended.

But lots of media are saying that there's a leaked report. Here's the HSJ's version (probably unpaywalled because of the outbreak):
https://www.hsj.co.uk/acute-care/exclusive-hospitals-70pc-more-full-than-april/7028742.article

Jonathan

[Ben@Forest]

According to the Telegraph (quoting an NHS update) there is not currently a higher occupation of intensive care beds than would be normal at this time of year.

TTBOMK that dataset is currently suspended.

But lots of media are saying that there's a leaked report. Here's the HSJ's version (probably unpaywalled because of the outbreak):
https://www.hsj.co.uk/acute-care/exclusive-hospitals-70pc-more-full-than-april/7028742.article

I know you live and die by the facts from the sources you prefer! But my wife is a nurse in NE England. She's not as busy as she was in the spring, leave has not been cancelled as it was in the spring.

[Oldjohnw]

And my niece is a doctor in Manchester and it's getting every bit as bad. Not quite but well on the way.

But the evidence of one person's niece and another's wife is not exactly statistically significant, although real.

I imagine we would rather arrest the thing before it was as bad as in the Spring rather than wait until it was as bad.

[reohn2]

And my niece is a doctor in Manchester and it's getting every bit as bad. Not quite but well on the way.

But the evidence of one person's niece and another's wife is not exactly statistically significant, although real.

I imagine we would rather arrest the thing before it was as bad as in the Spring rather than wait until it was as bad.

And not forgetting that we are entering winter,the flu season,which will,by all accounts,only add to the problems the country faces with C19.

[Ben@Forest]

But the evidence of one person's niece and another's wife is not exactly statistically significant, although real.

Exactly, I agree. I do think tho reports report the worst and the most vivid prognostications of statistics are chosen, not the whole.

Yesterday l was listening to a news report which told us a third of people were seriously concerned or worried about getting Covid. That means two thirds aren't. That's not to downplay the seriousness of the virus but it does say something about people's outlook and attitude. Two-thirds are getting on, being hopeful and will take it as it comes. That's a different slant to one-third are fearful.

[thirdcrank]

But the evidence of one person's niece and another's wife is not exactly statistically significant, although real.

Exactly, I agree. I do think tho reports report the worst and the most vivid prognostications of statistics are chosen, not the whole.

Yesterday l was listening to a news report which told us a third of people were seriously concerned or worried about getting Covid. That means two thirds aren't. That's not to downplay the seriousness of the virus but it does say something about people's outlook and attitude. Two-thirds are getting on, being hopeful and will take it as it comes. That's a different slant to one-third are fearful.

Not so long ago you had to agree that you had unintentionally misquoted data about something else.

Saying a third of people are seriously concerned doesn't tell us much about the other two thirds without knowing their responses to what is in any case a subjective question.

[Ben@Forest]

But the evidence of one person's niece and another's wife is not exactly statistically significant, although real.

Exactly, I agree. I do think tho reports report the worst and the most vivid prognostications of statistics are chosen, not the whole.

Yesterday l was listening to a news report which told us a third of people were seriously concerned or worried about getting Covid. That means two thirds aren't. That's not to downplay the seriousness of the virus but it does say something about people's outlook and attitude. Two-thirds are getting on, being hopeful and will take it as it comes. That's a different slant to one-third are fearful.

Not so long ago you had to agree that you had unintentionally misquoted data about something else.

Saying a third of people are seriously concerned doesn't tell us much about the other two thirds without knowing their responses to what is in any case a subjective question.

True. But this is an arguing about the stats thread. And that was a BBC report. Should such reports simply always be discounted as being too vague?

[simonineaston]

OK so what you got to remember is that nobody really knows anything nowadays... its like a zoom lens - zoom out a lot a and you can't see any detail. Zoom in too far, and you see a lot of detail but miss the big picture. Oh Dear. There simply isn't anybody who can see it all clearly. Except god of course... But don't ask me which one...

[thirdcrank]

... True. But this is an arguing about the stats thread. And that was a BBC report. Should such reports simply always be discounted as being too vague?

All I'm saying is that care needs to be taken with the interpretation of data, especially when interpreting somebody else's interpretation of data. When the data are responses to subjective questions, then it's a minefield. Before any question arises of a rigged poll, there's the real possibility of unintentional bias. eg Consider how any unsuspecting student coming on here with a survey generally gets an unsolicited short course in organising a survey.
=====================================================================
PS

viewtopic.php?f=7&t=142077

I rest my case

[simonineaston]

Oh - I forgot about that nice Doctor fellow - don't go mixing him up with god, mind...

[Ben@Forest]

... True. But this is an arguing about the stats thread. And that was a BBC report. Should such reports simply always be discounted as being too vague?

All I'm saying is that care needs to be taken with the interpretation of data, especially when interpreting somebody else's interpretation of data. When the data are responses to subjective questions, then it's a minefield. Before any question arises of a rigged poll, there's the real possibility of unintentional bias. eg Consider how any unsuspecting student coming on here with a survey generally gets an unsolicited short course in organising a survey.

I agree. It reminds me of the joke about the man whose family send him to a psychiatrist because he's fond of currant buns. 'Ridiculous' says the psychiatrist, 'I'm rather partial to currant buns myself'. Man says, 'Wonderful, you must come and visit, I've got a roomful'

We don't know where people's perceptions of 'concerned' or 'seriously worried' start. It will differ. But why did the BBC not say 66% are not concerned or seriously worried? Because our news reporting is based around the worst and the gloomiest. It sells better.