Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-L] void problem for 10 days + entertainment response

Seems to me the critical factor is localized population density--smaller areal choices than by county. While very much at risk for various reasons, nursing home/assisted care facilities have large population densities and by their very nature feature lots of person to person interactions. Not a surprise to me how big city life-styles like New York City are having such wide-spread infections. More familiar with Chicago, but the 'on the street' culture there seems somewhat different. (Bottom line, IMO. is that there are too many variables to expect well behaved statistics.)

Richard W. Tarara
Professor of Physics, emeritus
Saint Mary’s College
Notre Dame, Indiana

Free Physics Instructional Software
Windows and Mac

-----Original Message-----
From: Phys-l <> On Behalf Of Don via Phys-l
Sent: Tuesday, May 5, 2020 3:21 PM
Cc: Don <>
Subject: Re: [Phys-L] void problem for 10 days + entertainment response

I live in Anne Arundel County, MD and so was interested in this exchange. I obtained similar data to Brian Whatcott for all 24 Maryland counties (U.S. Census population data estimated to 2018 and Maryland Health Dept. Covid-19 case data as of 5/5/20). Plotting the data as county confirmed cases/100,000 versus county population size clearly shows an upward trend for the larger county population sizes. Overall correlation coefficient was about 0.7 indicating positive correlation with population size. R_sq for best fit straight line was only about 0.5 caused in part (I think) by "noisy" data at small population sizes. For counties with populations over 200,000, upward trend was quite clear (although not necessarily linear as the R_sq shows). However, for smaller population sizes, the county confirmed cases bounced around quite a bit making a trend with population size difficult to see.

Don Polvani

-----Original Message-----
From: Phys-l [] On Behalf Of
brian whatcott
Sent: Monday, May 4, 2020 2:00 PM
Subject: Re: [Phys-L] void problem for 10 days + entertainment

On 5/4/2020 6:46 AM, Bill Norwood via Phys-l wrote:
And, just for entertainment, since no numbers can be substantiated
compared: Do a log-log plot of the Maryland county by county confirmed
cases vs the county’s population.

You will see that the most highly populated counties (and cities)
have the
lowest numbers of confirmed cases per 10,000 population.

Issue: Should that be per 100/1000/10,000/100,000 population?

Pic of my plot available to individuals.

Bill Norwood

Sent from my iPhone

I set a list of Maryland counties ordered by size (US Census data)
against cases (Maryland Dept Health).
The smaller counties trended at 280 cases per hundred thousand county
population, shown here: small Maryland counties and big county
populations shown here: large Maryland counties The more populous
counties trended at 540 cases per hundred thousand county population.

Accordingly, I was unable to confirm your correlation, because I found the
highest populated counties showed the highest case rates. Comments?

Brian W
p.s. My data CAN be substantiated:
County case data:
County population data:
Forum for Physics Educators

Forum for Physics Educators