k2aggie07 said:

---

If they get 6% positive tests on 1800, that is 108 positives.

With 88.66% sensitivity and 90.63% specificity you'd expect this curve for true prevalence:



So either they are using a different sensitivity / specificity number or the 6% raw rate isn't correct. Shrug.

---

Quote:

---

Using statistical methods that account for the limitations of the test (sensitivity and specificity), we are 95% certain that the true amount of infection lies between 4.4% and 7.9% of the population, or between 123,000 and 221,000 residents.

---

k2aggie07 said:

---

My graph was wrong, I used 108 and 1800, should have been 108 and 1692. Changes it only a bit. Using the validation numbers from their website you liked, here's the more thorough estimate:



But how do you know that's the test kit they used? I can't find any information about which.

---