Pages: 1 ... 82 83 [84] 85 86 ... 126   Go Down

Author Topic: COVID-19 | science, damage limitation, NO POLITICS  (Read 197099 times)

PeterAit

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4647
    • Peter Aitken Photographs
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1660 on: July 05, 2020, 09:53:15 am »

And to cheer us all up, we now seem to have a covid mutation that makes it more likely to spread.

https://www.cnn.com/2020/07/02/health/coronavirus-mutation-spread-study/index.html

Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4600
    • Pieter Kers
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1661 on: July 06, 2020, 09:02:21 am »

The new normal
Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4600
    • Pieter Kers
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1662 on: July 07, 2020, 01:29:49 pm »

Informative graphic about the state of Covid-19in the world
Logged

hogloff

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1194
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1663 on: July 07, 2020, 04:15:46 pm »

Informative graphic about the state of Covid-19in the world

I wish the scale was the same for all graphs.
Logged

Alan Klein

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 15850
    • Flicker photos
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1664 on: July 07, 2020, 04:34:57 pm »

I wish the scale was the same for all graphs.
I agree.  Lies, damn lies, and statistics.

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8919
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1665 on: July 07, 2020, 05:43:40 pm »

I agree.  Lies, damn lies, and statistics.

How are they lies?

In fact, the benefit of the given choice of scaling is that it more easily identifies those who are performing sub-par.
Logged
== If you do what you did, you'll get what you got. ==

Robert Roaldi

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5511
    • Robert's Photos
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1666 on: July 07, 2020, 09:21:01 pm »

I agree.  Lies, damn lies, and statistics.

You have quoted that homily about lies and statistics many many times, you must like it.

But I heard a good line the other day. Yes, it's possible to lie using statistics, but it's a LOT easier to lie without them.
Logged
--
Robert

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4600
    • Pieter Kers
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1667 on: July 07, 2020, 10:24:45 pm »

I wish the scale was the same for all graphs.

I agree - here it is.   
Alan, what you see in those graphs is just that in some countries they test more than in others... ;)

btw some days later ; new cases july 7
africa 17.000 
Europe 12.000
Asia & Middle east 50.000
LatinAmerica & Carabbean 64.000
North America 64.000
« Last Edit: July 07, 2020, 10:49:27 pm by kers »
Logged

Alan Klein

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 15850
    • Flicker photos
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1668 on: July 08, 2020, 12:41:14 am »

You have quoted that homily about lies and statistics many many times, you must like it.

But I heard a good line the other day. Yes, it's possible to lie using statistics, but it's a LOT easier to lie without them.
But when you "lie" with statistics, it's more likely people will believe your lie.  After all, don't the statistics prove the point better than just words?  That's what makes statistics so dangerous. 

Ever read a forum argument comparing which camera lens is sharper?  Everyone has a statistic to prove their point.  All the pixel peepers come out claiming they can count more angels on the head of a pin with their lenses than any other.  They prove their point with charts, and pictures, and whatever. It's quite amusing I'm sure you'll agree. :)

Alan Klein

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 15850
    • Flicker photos
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1669 on: July 08, 2020, 12:44:07 am »

I agree - here it is.   
Alan, what you see in those graphs is just that in some countries they test more than in others... ;)

btw some days later ; new cases july 7
africa 17.000 
Europe 12.000
Asia & Middle east 50.000
LatinAmerica & Carabbean 64.000
North America 64.000
I suppose that's better Pieter.  But what does a "seven day rolling average" mean?

BobShaw

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2271
    • Aspiration Images
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1670 on: July 08, 2020, 01:44:08 am »

But when you "lie" with statistics, it's more likely people will believe your lie.  After all, don't the statistics prove the point better than just words?  That's what makes statistics so dangerous. 

I don't know about lying but you have to a bit deeper than the graph.
Asia for example is the largest continent on the planet with about half the worlds population, but it had less cases yesterday than the USA.
What does "North America" mean if it doesn't include Latin America and the Caribbean.
To me hard numbers trump a graph any day.
Logged
Website - http://AspirationImages.com
Studio and Commercial Photography

Alan Klein

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 15850
    • Flicker photos
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1671 on: July 08, 2020, 02:07:55 am »

I don't know about lying but you have to a bit deeper than the graph.
Asia for example is the largest continent on the planet with about half the worlds population, but it had less cases yesterday than the USA.
What does "North America" mean if it doesn't include Latin America and the Caribbean.
To me hard numbers trump a graph any day.
I agree.  How can you have any faith in a graph that doesn;t give all the parameters?  Like I asked, what's a 7 day rolling average when the graph extends for many months?   Where's South America?  Where the web link?  Who made the graph?  What does number of cases mean (from testing?  hospitalized?  etc.?)

jeremyrh

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2514
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1672 on: July 08, 2020, 02:59:17 am »

I agree.  How can you have any faith in a graph that doesn;t give all the parameters?  Like I asked, what's a 7 day rolling average when the graph extends for many months?   Where's South America?  Where the web link?  Who made the graph?  What does number of cases mean (from testing?  hospitalized?  etc.?)

It’s not the fault of the authors of the graph that you don’t understand words. Google will quickly tell you what a “seven day rolling average” is. Latin America is where they speak Latin.
Logged

jeremyrh

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2514
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1673 on: July 08, 2020, 03:40:03 am »

What does "North America" mean if it doesn't include Latin America and the Caribbean.


Umm ...  at a guess it means USA + Canada.
Logged

Paulo Bizarro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 7522
    • http://www.paulobizarro.com
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1674 on: July 08, 2020, 05:12:40 am »

I agree.  Lies, damn lies, and statistics.

This is ridiculous. Independent of the scale, the cases per region are what they are. For example, the USA would still be 50,000 cases a day.

BobShaw

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2271
    • Aspiration Images
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1675 on: July 08, 2020, 07:22:07 am »

Umm ...  at a guess it means USA + Canada.
Exactly. And as one of those is at number 1 and one is at number 20 that probably just means the US. Why not say that.
Logged
Website - http://AspirationImages.com
Studio and Commercial Photography

jeremyrh

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2514
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1676 on: July 08, 2020, 07:28:41 am »

Exactly. And as one of those is at number 1 and one is at number 20 that probably just means the US. Why not say that.

It's a comparison of figures per continent, apparently.  I think we all know what a basket case the US is in this regard!
Logged

rabanito

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1587
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1677 on: July 08, 2020, 07:58:01 am »

yes, in some countries. But increasing in most. Global death count for today: 2064
By country:
Italy - 743
USA - 120
Spain - 489
France - 240
Iran - 122
UK - 87
Netherlands - 63
None in Brazil?
Logged

Alan Goldhammer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4349
    • A Goldhammer Photography
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1678 on: July 08, 2020, 08:59:53 am »

The numbers are what they are.  One can use any measure to show the US response has been dismal.  Look at hospitalization usage in the current hard hit states.  In some regions they are at 100% capacity.  We know that there are improved treatments now and all those people will not die.  However, as was the case in New York, it puts a huge burden on the healthcare system and those who might need hospitalization for some non-COVID condition are put in peril.

One of the things I am tracking in my newsletter is the trend in infection and Case Rate Fatality (CRF).  CRF has traditionally been used by the public health community to gauge how problematic an epidemic is/was.  At present, it is an imprecise number as the total number of infections is not known with precision. I look at the numbers daily to see what the trend might be. The US CRF stands at 4.3%. I continue to believe as more accurate numbers come in this will drop by about a factor of 10 but that is still higher than any other influenza outbreak other than the Spanish Flu of 1918-19.

One can do the thought experiment based on herd immunity. In the absence of a vaccine, we will need to reach the herd immunity point which is unknowable right now. To make the math easy, assume it to be 60% of the population getting infected and recovering. That is 198 M people and we are no where close to that. If you assume my CFR estimate (0.5%) total mortality ends up at 990 K deaths, totally unacceptable which is why we need to be doing so much more.
Logged

Alan Klein

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 15850
    • Flicker photos
Re: COVID-19 | science, damage limitation, NO POLITICS
« Reply #1679 on: July 08, 2020, 09:20:52 am »

This is ridiculous. Independent of the scale, the cases per region are what they are. For example, the USA would still be 50,000 cases a day.
The shape of the original graphs were not the same distorting their relative meaning to the viewer projecting a different image.  Better if it just listed the totals in numbers than (edit) use fake graphs.  If you want to show volatility or amplify change, for example, just show the top 10% of a graph rather that the full 100% extent of the amounts.  It's an old graph trick.   
Pages: 1 ... 82 83 [84] 85 86 ... 126   Go Up