Georgia: Not out of the woods, and far from herd immunity

Below is a screenshot of testing data as of today in Georgia:

The percent positivity of PCR tests remains too high at over 6%. Anything above 5% indicates that either …

  • not enough tests are being performed to perform contact tracing, the lack of which could very well eventually allow the disease to become more prevalent; or
  • the contagion is spreading rapidly.

In addition, an antibody seropositive rate of only 8.5% indicates that Georgia is nowhere close to herd immunity.

This examination of the testing data is not meant to single out Georgia as being better or worse than any of the other 49 states.



Would you like to justify your title “far from herd immunity?”

How can you make such a statement without extensive data on who was tested and what a random sampling of tests would have revealed?

If the ground truth were that 50% of the population was seropositive, it would be well-nigh impossible to find 306,927 Georgians by any sampling method that would have 8.5% seropositivity.

This is Bayesian thinking, of course. When we lack the information needed to perform a frequentist analysis, Bayesian thinking gives us an imperfect but nonetheless very useful way to process the evidence we have.


I disagree with your assessment of “well-nigh impossible.”

There are more than 10 million people in Georgia.

Only the curious wealthy or those working in health care will typically take an antibody test.

You are free to your opinion. I find your claim of “far from herd immunity” as completely unsupported.

I suspect the antibody tests are far, far from a random sampling.

COVID-19 herd immunity: where are we?

Nature Reviews Immunology volume 20 , pages 583–584(2020)

Herd immunity is achieved when one infected person in a population generates less than one secondary case on average …

When you see “super-spreader” events still happening, such as the WH Rose Garden ceremony, then you know we aren’t anywhere near herd immunity.

Assuming an optimistic herd immunity threshold of 50%, for countries such as France and the USA, this would translate into 100,000–450,000 and 500,000–2,100,000 deaths, respectively.

Simple math: 10% presently immune and 210,000 dead multiplied by 5 = 50% immune (optimistic herd immunity) and 1,050,000 dead. Is that a price worth paying? I don’t think so.


3% of the population is an extremely large sample. Election polls typically have a margin of error of about 2-3% with a sample size of only 0.001% of the registered voters

1 Like

We were discussing Georgia, not the White House nor the planet.

See the title of the thread.

1 Like

Chris, you don’t seem to understand that the 3% that have taken an antibody test do not represent a “sample” of the state.

It is a specific subset, not a sample.

I don’t understand why you don’t realize that.

Make as many snappy putdowns on behalf of frequentism as you like, Vance, but I doubt you will convince many readers with such a barrage.

Health care workers are among the most afflicted by COVID due to their necessary exposure to spreaders and the ongoing lack of personal protective equipment in too many settings. Thus you are making a very strong argument that the seropositivity level in the 3% sample (and that is indeed the correct word) is probably higher than it is in the general population.

A sample does not need to be random to be a sample.


You really are hopeless. Maybe I’ll check back on you in the Spring, assuming both of us survive this Winter.


If you think noting a specific subset is not a sample is a snappy comeback, then I am guilty.

But I understand your posts to be misleading and inaccurate.

On the contrary, I am optimistic and hopeful.

1 Like

Regarding this, I was on another site for physicians, and the question came up if they were being tested by either PCR, rapid molecular testing, or antibody testing. Most are in large organizations that have mandated protocols to follow regarding testing. It was interesting to see how the policies were all over the chart, from no testing if asymptomatic, to twice weekly swabs, and occ. antibody tests.

The consensus was that antibody tests were useless for the individual, and helpful only in the long run to researchers. The results are highly dependent on the subgroup you are testing. If you are doing antibody testing of people who had confirmed PCR positive disease, it is pretty indicative of an immune response. However, a fair number of folks turn negative for antibodies in a couple of months, making it a poor indicator of past infection, and of course leading to questions as to how long lasting immunity will be. If done in mass screening of people who are not known to have disease, a large percentage of the positives will be false positives, again making interpretation of results problematic. That is why we have epidemiologists, I suppose, who have to training and knowledge to ultimately tease something meaningful out of the raw data.


My wife and I had antibody tests in April because we were on a tour in early March with someone who tested positive for Covid-19, and we passed through JFK airport on the way back from the tour.

We were interested in whether or not we had contracted the disease.

I think the overall results for the state of Georgia on antibody testing tells us little, since the testing data base is not close to a sample that reflects the population in general.

Frequentists commonly hurl this accusation at Bayesians.


Sorry, but it’s a sample. Specifically, it’s a nonrandom ‘convenience sample’.

1 Like

Can we leave out melodrama and just stick to facts? I enjoy these on the level of #1 targeted #2 scientific #3 respectful #4 no melodrama, which is akin to a strawman approach used to divert the argument to sensationalism.

1 Like

Do you think it shows what the whole is like? If so, why?

Hi and welcome. Sorry, but Vance has a long history on the subject that you likely missed, and to which I was responding. He has his own concepts of herd immunity that no amount of facts has breached. I wasn’t being melodramatic when I said I’d check back in the Spring, because he won’t concede on anything until the facts are known, which won’t be until Winter is over and we know the numbers. As for surviving, that’s not melodrama either. Taking comorbidities into account, I give myself a 50% chance if I contract the virus. It’s highly contagious and my wife works in COVID medical research, so I don’t have a good feeling. We’ll see what how it all plays out come Spring.

Stay safe!

1 Like

I haven’t seen anything on this thread that justifies the claim of “far from herd immunity,” which is in the title of this thread.

But I am still watching.

Antibody test totals with no information on who took the tests does not rise to the level of evidence or proof.

“Let your conversation be always full of grace, seasoned with salt, so that you may know how to answer everyone.” -Colossians 4:6

This is a place for gracious dialogue about science and faith. Please read our FAQ/Guidelines before posting.