Replies: 37
| visibility 1,269
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Okay CDC says mask mandates are effective
Mar 8, 2021, 9:47 PM
|
|
|
|
|
|
Hall of Famer [24289]
TigerPulse: 100%
Posts: 13874
Joined: 7/3/01
|
Irrelevant. Election was in November.
Mar 8, 2021, 10:01 PM
|
|
Impressive number of words and symbols, too, Not far from eight by ten glossy photographs with circles and arrows and a paragraph on the back of each one.
Message was edited by: CUintulsa®
|
|
|
|
|
All-TigerNet [10776]
TigerPulse: 100%
Posts: 14958
Joined: 8/6/10
|
LOL, they are so full of sh*t! haha***
Mar 8, 2021, 10:06 PM
|
|
|
|
|
|
|
Lot o points [154409]
TigerPulse: 100%
Posts: 65237
Joined: 5/6/13
|
yeah, but the infographic!!!!***
Mar 8, 2021, 10:14 PM
|
|
|
|
|
|
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Re: yeah, but the infographic!!!!***
Mar 8, 2021, 10:21 PM
|
|
|
|
|
|
|
Oculus Spirit [97320]
TigerPulse: 100%
Posts: 64579
Joined: 7/13/02
|
It affected the r or growth rate by that number.
Mar 9, 2021, 5:13 AM
|
|
Not total cases.
But by all means... Carry on.
|
|
|
|
|
All-In [30923]
TigerPulse: 100%
Posts: 9982
Joined: 1/28/15
|
^^^
Mar 9, 2021, 5:49 AM
|
|
Crickets
|
|
|
|
|
All-In [40797]
TigerPulse: 100%
Posts: 42891
Joined: 11/30/98
|
why say crickets instead of pointing out that he is
Mar 9, 2021, 6:23 AM
|
|
incorrect. Growth rate is not R. If R changed by 1.8 that would be a big deal and easily noticed. Clearly this is not the case.
|
|
|
|
|
Team Captain [499]
TigerPulse: 68%
Posts: 803
Joined: 11/30/98
|
|
|
|
|
All-In [40797]
TigerPulse: 100%
Posts: 42891
Joined: 11/30/98
|
Masks help
Mar 9, 2021, 8:45 AM
|
|
but that doesn't mean you destroy business in the name of helping.
And of course the "worries" of the experts were unfounded. The fear they were trying to spread was typical.
|
|
|
|
|
All-In [30923]
TigerPulse: 100%
Posts: 9982
Joined: 1/28/15
|
|
|
|
|
All-In [46664]
TigerPulse: 100%
Posts: 30614
Joined: 8/11/15
|
Listen here you little ####. I spent tens of minutes
Mar 10, 2021, 9:46 AM
|
|
googling this.
I don't need your 8 year medical schooling to tell me something that goes against what I feel.
--T3, probably
Message was edited by: FBCoachSC®
|
|
|
|
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Re: Listen here you little ####. I spent tens of minutes
Mar 10, 2021, 6:42 PM
|
|
|
|
|
|
|
Lot o points [154409]
TigerPulse: 100%
Posts: 65237
Joined: 5/6/13
|
Help me out here
Mar 10, 2021, 10:06 AM
[ in reply to It affected the r or growth rate by that number. ] |
|
It's been 20+ years since my last statistics class, so I could be missing something, but.....
they define growth rate as: "The daily growth rate was defined as the difference between the natural log of cumulative cases or deaths on a given day and the natural log of cumulative cases or deaths on the previous day, multiplied by 100"
Unless I'm just whiffing on this, a 1.8% increase in the growth rate correlates completely with the total number of cases on a given day being 1.8% higher than the day before.
I could easily be wrong here, so if I am, please explain it to me like I'm 15.
|
|
|
|
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Re: Help me out here
Mar 10, 2021, 4:05 PM
|
|
|
|
|
|
|
Hall of Famer [22722]
TigerPulse: 100%
Posts: 16813
Joined: 12/2/00
|
Log transformation is popular when looking at skewed numbers
Mar 10, 2021, 11:37 PM
[ in reply to Help me out here ] |
|
It helps to fit the data to a more normal distribution (think bell shape curve). This is a common technique and helps when evaluating data sets.
|
|
|
|
|
All-In [31784]
TigerPulse: 100%
Posts: 37085
Joined: 11/22/03
|
It's is worded really confusingly....
Mar 11, 2021, 8:03 AM
[ in reply to Help me out here ] |
|
if they're using the term "growth rate" correctly and it's expressed as "daily"...that means that each day's rate of growth was 1.8% less mask mandate vs no mandate. So, after 30 days, the mask mandate would have 1.8 * 30 = 54% less absolute cases.
|
|
|
|
|
All-In [31784]
TigerPulse: 100%
Posts: 37085
Joined: 11/22/03
|
*it is****
Mar 11, 2021, 8:04 AM
|
|
|
|
|
|
|
Lot o points [154409]
TigerPulse: 100%
Posts: 65237
Joined: 5/6/13
|
Right but they don't seem to be doing it cumulatively
Mar 11, 2021, 8:42 AM
[ in reply to It's is worded really confusingly.... ] |
|
"During March 1–December 31, 2020, state-issued mask mandates applied in 2,313 (73.6%) of the 3,142 U.S. counties. Mask mandates were associated with a 0.5 percentage point decrease (p = 0.02) in daily COVID-19 case growth rates 1–20 days after implementation and decreases of 1.1, 1.5, 1.7, and 1.8 percentage points 21–40, 41–60, 61–80, and 81–100 days, respectively, after implementation (p<0.01 for all) (Table 1) (Figure). Mask mandates were associated with a 0.7 percentage point decrease (p = 0.03) in daily COVID-19 death growth rates 1–20 days after implementation and decreases of 1.0, 1.4, 1.6, and 1.9 percentage points 21–40, 41–60, 61–80, and 81–100 days, respectively, after implementation (p<0.01 for all). Daily case and death growth rates before implementation of mask mandates were not statistically different from the reference period.
It seems like they're looking at a non-mandate location vs a mandate location and examining the growth rate difference on a single day's snapshot on day 20, 40, 60, etc.....So it's not really accounting for the ups and downs in a month as much as that single day view when compared.
Fully admit again I could be wrong, as we all agree the wording on this is seriously confusing, but these numbers don't show be that a mask mandate passes the risk/reward test at all. I suspect we could overlay these results to flu infection/death rates (something we've lived with and accepted the casualties from for quite a long time) and get similar results.
|
|
|
|
|
All-In [31784]
TigerPulse: 100%
Posts: 37085
Joined: 11/22/03
|
Yep...as I said above somewhere...
Mar 11, 2021, 8:59 AM
|
|
this study would be much more meaningful if they put the numbers in or made a model in absolute case count differences so the reader could have some context.
But growth rate has to be/should be cumulative, otherwise, why make it a rate?
You mentioned risk/reward for a mask mandate. What is the risk side of a mask mandate?
|
|
|
|
|
Lot o points [154409]
TigerPulse: 100%
Posts: 65237
Joined: 5/6/13
|
Sorry, I only referenced that but was referring to both
Mar 11, 2021, 9:15 AM
|
|
masks and in-person dining restrictions that the study focused on. Clearly the risks are economic.
|
|
|
|
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
|
|
|
|
Hall of Famer [22722]
TigerPulse: 100%
Posts: 16813
Joined: 12/2/00
|
Re: Yep...as I said above somewhere...
Mar 11, 2021, 4:04 PM
|
|
R-squared measures how much variation can be explained by your model. By changing one value, how much is expected to change in the other. It's a 'squared' relationship, so a .5 R value is a .25 R-squared value
P-value measures the significance. Get that less than 0.05 and it's statistically significant.
|
|
|
|
|
All-In [31784]
TigerPulse: 100%
Posts: 37085
Joined: 11/22/03
|
I don't think you're reading those stats correctly...
Mar 9, 2021, 7:46 AM
|
|
growth rate is not the same as # of cases.
Using growth rate instead of percent change is admittedly confusing. I know it's how studies like this are normally measured, but it makes it hard for people not used to dealing with statistics to interpret the data. To make it worse, they used logarithmic (natural log) data.
|
|
|
|
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Re: I don't think you're reading those stats correctly...
Mar 9, 2021, 10:49 AM
|
|
|
|
|
|
|
All-In [31784]
TigerPulse: 100%
Posts: 37085
Joined: 11/22/03
|
Well, the fact of the matter is this thing, in it's...
Mar 9, 2021, 11:58 AM
|
|
current form has only been around about 1 year. That isn't much time to be able to answer some of the types of questions you posed. As for why some hard than others, most of what I've read seems to point to DNA/genetics.
I have a hard time gauging the practical impact of 2% reduction in rate of growth. I wish they would translate that back into a case count model for context. I know it's not that easy with this kind of data that has a time component, but not many are going to be able to interpret the impact of masks by this data. I think most are going to come away believing masks just reduced the number of covid cases by 2% and think big whoop.
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
Most people will never read this the way we are.
Mar 10, 2021, 8:34 AM
|
|
Most people will just use someone else's Facebook opinion as a basis for their own opinion, in my experience.
|
|
|
|
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Re: Most people will never read this the way we are.
Mar 10, 2021, 11:31 AM
|
|
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
|
|
|
|
All-In [26968]
TigerPulse: 96%
Posts: 44823
Joined: 7/6/10
|
Excellent!***
Mar 10, 2021, 8:31 AM
|
|
|
|
|
|
|
All-In [26309]
TigerPulse: 100%
Posts: 20513
Joined: 9/2/02
|
and who pays the price for this?
Mar 10, 2021, 9:43 AM
|
|
All the little sea turtles with disposable masks wrapped around their necks
|
|
|
|
|
Hall of Famer [22722]
TigerPulse: 100%
Posts: 16813
Joined: 12/2/00
|
We talking percents or percentage points?
Mar 10, 2021, 9:00 PM
|
|
1.8 percentage points is quite large. Pretty sure they’re referencing percentage points.
For example, changing a growth from 10.0 to 11.8 percentage points is an 18% difference.
Changing growth from 5.0 to 6.8 is a 36% difference.
I didn’t read the detail enough to see what the starting growth rates were.
|
|
|
|
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Re: We talking percents or percentage points?
Mar 10, 2021, 9:18 PM
|
|
|
|
|
|
|
Hall of Famer [22722]
TigerPulse: 100%
Posts: 16813
Joined: 12/2/00
|
Re: We talking percents or percentage points?
Mar 11, 2021, 12:02 AM
|
|
The sentence before that and every statement after references percentage points
“Two outcomes were examined: the daily percentage point growth rate of county-level COVID-19 cases and county-level COVID-19 deaths. “
The math works as taking ln(A) versus ln(B), then assessing percent growth. Add 1.8% to that percent growth. So a 4.0% growth would become a 5.8% growth. As mentioned by somebody else, this only measured the impact of the mandates and did not measure actually wearing a mask.
Changing growth rate of 1.8 points with just a mandate is HUGE. If at peak the case fatality rate in US was 6.0%, then a 1.8 point change in that rate is extraordinary.
|
|
|
|
|
Rock Defender [52]
TigerPulse: 90%
Posts: 35
Joined: 11/30/98
|
Re: We talking percents or percentage points?
Mar 11, 2021, 1:09 AM
|
|
|
|
|
|
|
Hall of Famer [22722]
TigerPulse: 100%
Posts: 16813
Joined: 12/2/00
|
These articles remind me
Mar 11, 2021, 4:13 PM
|
|
how much I hate reading when stats people write articles. Regardless, I've enjoyed some of the research. Thanks.
|
|
|
|
|
All-In [40797]
TigerPulse: 100%
Posts: 42891
Joined: 11/30/98
|
|
|
|
|
Hall of Famer [22722]
TigerPulse: 100%
Posts: 16813
Joined: 12/2/00
|
Re: a 1.8% increase on 4% makes the 4% go to 4.072%***
Mar 11, 2021, 4:12 PM
|
|
A 1.8% increase on 4% makes the 4% go to 4.072%.
A 1.8% point increase on 4% makes the 4% go to 5.8%
Two wildly different results especially if looking at growth difference. It's flawed in stats work to do a percent on a percent, and the standard is to look at Point difference when comparing to percents.
|
|
|
|
Replies: 37
| visibility 1,269
|
|
|