NYU Professor Creates COVID-19 Dashboard to Compare Country and State Data

[Tom Johnson]

NYU Professor Creates COVID-19 Dashboard to Compare Country and State Data

by Sabrina I. Pacifici on Jan 26, 2021

“A new online dashboard, created by NYU Professor Alexej Jerschow, brings together COVID-19 data from U.S. states and countries around the world to compare cases, deaths, vaccines, and testing in a visual, user-friendly format.  The tool also integrates a range of policies governments have implemented to limit the spread of COVID-19—including school closings, stay-at-home orders, and mask mandates—in an effort to compare policy responses with COVID-19 outcomes.Jerschow, who is a professor of chemistry and typically works in the field of magnetic resonance imaging and battery research, was looking for a convenient way to compare COVID-19 data from different countries and states. While the data were available from different sources, there was not a single tool that allowed him to easily analyze how different geographic areas were faring and responding to the pandemic—so, he created one…The dashboard combines data from a series of publicly available sources. Country-wide information is downloaded from Our World in Data, which sources its data from the Johns Hopkins Center for Systems Science and Engineering dashboard. U.S. state and territory data are obtained from covidtracking.com. The Oxford COVID-19 Government Response Tracker (OxCGRT) provides policy data…”

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: http://friam.471366.n2.nabble.com/

[Sarbajit Roy (testing)]

GIGO.

COVID-19 statistics (data) is suspect and not strictly comparable across countries, starting with China's numbers.

On Thu, Jan 28, 2021 at 1:20 AM Tom Johnson <[hidden email]> wrote:

NYU Professor Creates COVID-19 Dashboard to Compare Country and State Data

by Sabrina I. Pacifici on Jan 26, 2021

“A new online dashboard, created by NYU Professor Alexej Jerschow, brings together COVID-19 data from U.S. states and countries around the world to compare cases, deaths, vaccines, and testing in a visual, user-friendly format.  The tool also integrates a range of policies

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: http://friam.471366.n2.nabble.com/

[Barry MacKichan]

Tom, you’re a journalist, so now that I have your (virtual and asynchronous) attention, I’d like to register some complaints.

We seem to get the same data again and again, but there is critical data that noone has included in what I’ve read. Two examples:

1. We get daily death rates and number of hospitalizations, but no information about the average length of a hospital stay for a covid patient. This may not seem important, but if I want to compute the probability of dying from covid once you are hospitalized, I need it. It is (deaths/hospitalizations)*length-of-stay — this is because one person contributes 1 to the count of hospitalization for each day he is hospitalized, but can contribute at most 1 to the deaths. The naive calculation of the probability would be about 3% (3K/100K), but if the length of stay is 5 or 6 days, the true calculations show the odds of survival are the same as one round of Russian roulette.

It’s possible that the journalists have not given us this data because the data sites (covidtracking.com and Johns Hopkins) don’t have it. Good opportunity for a “scoop”.

2. A definition of a voting machine. Based on my experience in NM and in North Carolina, the only machines were scanners. I have seen one photograph of a Dominion “voting machine”, and the branding on the machine said “Dominion scanner.” It would be hard to corruptly modify the results from the scanner, and absolutely impossible to do so in a case where the hand count matches the scanner count. Fifteen years ago, there were lots of voting machines that did not produce paper ballots, and these were eminently hackable. But no journalist I’ve read has made any distinction between the machine types (and it is possible that all modern voting machines are scanners reading paper ballots. If so, tell us.)

You are of course innocent of these lapses, but maybe you could tell some of your journalist friends that they are leaving out vital information.

Thanks for listening (if you got this far).

—Barry

On 27 Jan 2021, at 14:49, Tom Johnson wrote:

NYU Professor Creates COVID-19 Dashboard to Compare Country and State Data

by Sabrina I. Pacifici on Jan 26, 2021

“A new online dashboard, created by NYU Professor Alexej Jerschow, brings together COVID-19 data from U.S. states and countries around the world to compare cases, deaths, vaccines, and testing in a visual, user-friendly format.  The tool also integrates a range of policies governments have implemented to limit the spread of COVID-19—including school closings, stay-at-home orders, and mask mandates—in an effort to compare policy responses with COVID-19 outcomes.Jerschow, who is a professor of chemistry and typically works in the field of magnetic resonance imaging and battery research, was looking for a convenient way to compare COVID-19 data from different countries and states. While the data were available from different sources, there was not a single tool that allowed him to easily analyze how different geographic areas were faring and responding to the pandemic—so, he created one…The dashboard combines data from a series of publicly available sources. Country-wide information is downloaded from Our World in Data, which sources its data from the Johns Hopkins Center for Systems Science and Engineering dashboard. U.S. state and territory data are obtained from covidtracking.com. The Oxford COVID-19 Government Response Tracker (OxCGRT) provides policy data…”

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: http://friam.471366.n2.nabble.com/

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: http://friam.471366.n2.nabble.com/

[Prof David West]

Barry,

Some dated information: a long way back, I was on a "blue ribbon committee" that included voting officials from six states in the Midwest plus several electronic engineers and several software engineers. The purpose was to evaluate voting machines and select a vendor for the six states. Dominion was one of the competing vendors.

None of the machines were simple scanners. All had software for tabulation, error correction, statistical analysis of various kinds, real-time counts, remote monitoring over a network, etc.

A mandate from the election officials — all machines had to have a means for monitoring and "correcting" results, i.e. to let election officials override what the machines were tabulating. This override mechanism was semi-secret, but like I said it was mandated by the RFP and so it was present in all the machines we evaluated then.

If the backdoor was used, a record of what the official did was recorded in the audit trail along with all other counting information.

Dominion was the only vendor that provided a way to 'turn off' the audit trail feature. As offered, at that time, election officials had the power to sit in their offices, watch the results, and "correct" them if deemed necessary, invisible to any subsequent audit checks.

Discrepancies between hand count and machine count were minimized because ballots were sorted and bundled, such that bundles, not individual ballots were actually counted if a recount was mandated.

The whole experience made me a non-believer in any electronic election.

All this may be quite different today — I only report what was.

davew

On Thu, Jan 28, 2021, at 8:38 AM, Barry MacKichan wrote:

Tom, you’re a journalist, so now that I have your (virtual and asynchronous) attention, I’d like to register some complaints.

We seem to get the same data again and again, but there is critical data that noone has included in what I’ve read. Two examples:

1. We get daily death rates and number of hospitalizations, but no information about the average length of a hospital stay for a covid patient. This may not seem important, but if I want to compute the probability of dying from covid once you are hospitalized, I need it. It is (deaths/hospitalizations)*length-of-stay — this is because one person contributes 1 to the count of hospitalization for each day he is hospitalized, but can contribute at most 1 to the deaths. The naive calculation of the probability would be about 3% (3K/100K), but if the length of stay is 5 or 6 days, the true calculations show the odds of survival are the same as one round of Russian roulette.
   

It’s possible that the journalists have not given us this data because the data sites (covidtracking.com and Johns Hopkins) don’t have it. Good opportunity for a “scoop”.

2. A definition of a voting machine. Based on my experience in NM and in North Carolina, the only machines were scanners. I have seen one photograph of a Dominion “voting machine”, and the branding on the machine said “Dominion scanner.” It would be hard to corruptly modify the results from the scanner, and absolutely impossible to do so in a case where the hand count matches the scanner count. Fifteen years ago, there were lots of voting machines that did not produce paper ballots, and these were eminently hackable. But no journalist I’ve read has made any distinction between the machine types (and it is possible that all modern voting machines are scanners reading paper ballots. If so, tell us.)
   

You are of course innocent of these lapses, but maybe you could tell some of your journalist friends that they are leaving out vital information.

Thanks for listening (if you got this far).

—Barry

On 27 Jan 2021, at 14:49, Tom Johnson wrote:

NYU Professor Creates COVID-19 Dashboard to Compare Country and State Data

by Sabrina I. Pacifici on Jan 26, 2021

“A new online dashboard, created by NYU Professor Alexej Jerschow, brings together COVID-19 data from U.S. states and countries around the world to compare cases, deaths, vaccines, and testing in a visual, user-friendly format.  The tool also integrates a range of policies governments have implemented to limit the spread of COVID-19—including school closings, stay-at-home orders, and mask mandates—in an effort to compare policy responses with COVID-19 outcomes.Jerschow, who is a professor of chemistry and typically works in the field of magnetic resonance imaging and battery research, was looking for a convenient way to compare COVID-19 data from different countries and states. While the data were available from different sources, there was not a single tool that allowed him to easily analyze how different geographic areas were faring and responding to the pandemic—so, he created one…The dashboard combines data from a series of publicly available sources. Country-wide information is downloaded from Our World in Data, which sources its data from the Johns Hopkins Center for Systems Science and Engineering dashboard. U.S. state and territory data are obtained from covidtracking.com. The Oxford COVID-19 Government Response Tracker (OxCGRT) provides policy data…”

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .

FRIAM Applied Complexity Group listserv

Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam

un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com

FRIAM-COMIC http://friam-comic.blogspot.com/

archives: http://friam.471366.n2.nabble.com/

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .

FRIAM Applied Complexity Group listserv

Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam

un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com

FRIAM-COMIC http://friam-comic.blogspot.com/

archives: http://friam.471366.n2.nabble.com/

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: http://friam.471366.n2.nabble.com/

[Tom Johnson]

Barry:

Thanks for taking the time to analyze and write.  
Of course, you are correct on both points.  My guess is that the problem/challenge on #1 is that the data on hospital stays is, at best, being collected only at the individual hospital and not yet being aggregated at any higher level.  But perhaps someone is this merry band knows otherwise.  Second, there is a culture of data hugging when it comes to hospitals, often driven by uninformed assumptions about  HIPAA.  So while such data probably exists, an individual hospital or hospital group are reluctant to give it out.  All that said, the problem with journalists and data is that the majority have at, best limited, interest in data and analyzing it.  Remember, journalism is the second refuge of the mathematically phobic.  One of the worst examples I see is every day in the New Mexican, where it carries some stats and data of the latest COVID count, but does so showing things like the latest change by ZIP.  (a) This fails to account for the vast differences in ZIP code population, hence (b) failure to put the cases in the context of cases per 1,000 or 10,000.

And yet again, the devices are not voting machines.  They are vote-counting machines, but the general public -- including journos -- don't know why or how they are different.

Thanks again,

Tom

On Thu, Jan 28, 2021 at 8:38 AM Barry MacKichan <[hidden email]> wrote:

Tom, you’re a journalist, so now that I have your (virtual and asynchronous) attention, I’d like to register some complaints.

We seem to get the same data again and again, but there is critical data that noone has included in what I’ve read. Two examples:

1. We get daily death rates and number of hospitalizations, but no information about the average length of a hospital stay for a covid patient. This may not seem important, but if I want to compute the probability of dying from covid once you are hospitalized, I need it. It is (deaths/hospitalizations)*length-of-stay — this is because one person contributes 1 to the count of hospitalization for each day he is hospitalized, but can contribute at most 1 to the deaths. The naive calculation of the probability would be about 3% (3K/100K), but if the length of stay is 5 or 6 days, the true calculations show the odds of survival are the same as one round of Russian roulette.

It’s possible that the journalists have not given us this data because the data sites (covidtracking.com and Johns Hopkins) don’t have it. Good opportunity for a “scoop”.

2. A definition of a voting machine. Based on my experience in NM and in North Carolina, the only machines were scanners. I have seen one photograph of a Dominion “voting machine”, and the branding on the machine said “Dominion scanner.” It would be hard to corruptly modify the results from the scanner, and absolutely impossible to do so in a case where the hand count matches the scanner count. Fifteen years ago, there were lots of voting machines that did not produce paper ballots, and these were eminently hackable. But no journalist I’ve read has made any distinction between the machine types (and it is possible that all modern voting machines are scanners reading paper ballots. If so, tell us.)

You are of course innocent of these lapses, but maybe you could tell some of your journalist friends that they are leaving out vital information.

Thanks for listening (if you got this far).

—Barry

On 27 Jan 2021, at 14:49, Tom Johnson wrote:

NYU Professor Creates COVID-19 Dashboard to Compare Country and State Data

by Sabrina I. Pacifici on Jan 26, 2021

“A new online dashboard, created by NYU Professor Alexej Jerschow, brings together COVID-19 data from U.S. states and countries around the world to compare cases, deaths, vaccines, and testing in a visual, user-friendly format.  The tool also integrates a range of policies governments have implemented to limit the spread of COVID-19—including school closings, stay-at-home orders, and mask mandates—in an effort to compare policy responses with COVID-19 outcomes.Jerschow, who is a professor of chemistry and typically works in the field of magnetic resonance imaging and battery research, was looking for a convenient way to compare COVID-19 data from different countries and states. While the data were available from different sources, there was not a single tool that allowed him to easily analyze how different geographic areas were faring and responding to the pandemic—so, he created one…The dashboard combines data from a series of publicly available sources. Country-wide information is downloaded from Our World in Data, which sources its data from the Johns Hopkins Center for Systems Science and Engineering dashboard. U.S. state and territory data are obtained from covidtracking.com. The Oxford COVID-19 Government Response Tracker (OxCGRT) provides policy data…”

============================================
Tom Johnson - tom@...
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
NM Foundation for Open Government
Check out It's The People's Data                 

============================================

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: http://friam.471366.n2.nabble.com/

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: http://friam.471366.n2.nabble.com/

- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: http://friam.471366.n2.nabble.com/