More Boris lockdown breaches

[BeElBeeBub]

Despite it stating clearly it wasn't.  You have to do better than that.  It's a disingenuous conversation if you can't admit it.

The only people who can know what the specificity in S2 was are the authors.
I will see if there is any way of getting clarification.

However, you still need to state what you think the figure was if not 99.9%. you obviously think.it was.lower.

Fine - what was it?

And then, having told.us what the figure was......say why you think the figure used in a sensitivity analysis, by definition not the value believed to be the actual value,  is important.

[Streetwalker]

Slightly more on topic:

The report is damning.

More importantly you can see, by his very trumpian reaction to it, exactly how much contempt he holds parliament in.

For a man who relied very heavily on the "sovereignty of parliament" he's showing his true colours.

Unfortunately with the current rabble that sit in Westminster contempt for Parliament is pretty much a given across the Kingdoms .
That we still believe in democracy is probably the only thing stopping rebellion .

[Scott777]

You want me to "admit specificity wasn't 99.9% for both scenarios".

I firmly believe it was.

Despite it stating clearly it wasn't.  You have to do better than that.  It's a disingenuous conversation if you can't admit it.

[BeElBeeBub]

I'm going to ignore everything in this post, because you still won't admit the specificity was not 99.9% for both scenarios.  Instead, you just want to dodge the facts.  That's gaslighting.  Have the decency to admit it.

You ignoring everything is very on brand.

You want me to "admit specificity wasn't 99.9% for both scenarios".

I firmly believe it was.

However, let's say I am wrong and you are right. The specificity in S2 was some other value than 99.9%

What value do you think it was?
How does the value chosen for a scenario affect what specificity was actually calculated to be?
How do you explain the observed results being consistent with your assertion specificity was lower than 99.9%?

[BeElBeeBub]

Slightly more on topic:

The report is damning.

More importantly you can see, by his very trumpian reaction to it, exactly how much contempt he holds parliament in.

For a man who relied very heavily on the "sovereignty of parliament" he's showing his true colours.

[Unlucky4Sum]

Anyway back on the thread-header subject:-

Matter referred on 21 April 2022 (conduct of Rt Hon Boris Johnson): Final Report (parliament.uk)

Our final conclusion is in relation to sanction. Although Mr Johnson's resignation as an MP renders it impossible for a sanction of suspension to be imposed, we draw attention to the fact that before the events of Friday 9 June we had provisionally agreed to recommend a suspension long enough to engage the provisions of the Recall of MPs Act. In the light of Mr Johnson's further contempts, we put on record that if he had not resigned his seat, we would have recommended that he be suspended from the service of the House for 90 days for repeated contempts and for seeking to undermine the parliamentary process, by: • Deliberately misleading the House • Deliberately misleading the Committee • Breaching confidence • Impugning the Committee and thereby undermining the democratic process of the House • Being complicit in the campaign of abuse and attempted intimidation of the Committee. In view of the fact that Mr Johnson is no longer a Member, we recommend that he should not be granted a former Member's pass

Finally he's called out on the record for being the complete dishonest dickhead some of us always saw him for. 

[papasmurf]

Anyway back on the thread-header subject:- 

Matter referred on 21 April 2022 (conduct of Rt Hon Boris Johnson): Final Report (parliament.uk)

Our final conclusion is in relation to sanction. Although Mr Johnson's resignation as an MP renders it impossible for a sanction of suspension to be imposed, we draw attention to the fact that before the events of Friday 9 June we had provisionally agreed to recommend a suspension long enough to engage the provisions of the Recall of MPs Act. In the light of Mr Johnson's further contempts, we put on record that if he had not resigned his seat, we would have recommended that he be suspended from the service of the House for 90 days for repeated contempts and for seeking to undermine the parliamentary process, by: • Deliberately misleading the House • Deliberately misleading the Committee • Breaching confidence • Impugning the Committee and thereby undermining the democratic process of the House • Being complicit in the campaign of abuse and attempted intimidation of the Committee. In view of the fact that Mr Johnson is no longer a Member, we recommend that he should not be granted a former Member's pass

[Scott777]

Welcome back.

1) what specificity rate do you think they are using for scenario 2?
2) given scenarios scenarios are just paper exercises to examine the sensitivity of the modeling to the input variables: why does the figure they pick have any bearing on what the actual figure is?

They could have created a scenario 3 where specificity was set to 50% That doesn't mean specificity was actually 50%, just "this is what the figures might look like if it was"

Your entire argument that the PCR specificity was *not* 99.9% or better is based around the fact the paper didn't explicitly (it did implicitly) define the specificity in one scenario.

Talking of ignoring: what is your response to the statistical fact that there is a 0.006% for a test with a specificity of less than 99.9% to give the observed results?

I'm going to ignore everything in this post, because you still won't admit the specificity was not 99.9% for both scenarios.  Instead, you just want to dodge the facts.  That's gaslighting.  Have the decency to admit it.

[BeElBeeBub]

You can just ignore the actal quotes I posted, and carry on making false claims in blissful ignorance, but it only exposes yourself.  Or, just admit you are wrong.  You had 2 weeks to think about it.

"we have estimated what prevalence would be in two scenarios using different possible test sensitivity and specificity rates."

Welcome back.

1) what specificity rate do you think they are using for scenario 2?
2) given scenarios scenarios are just paper exercises to examine the sensitivity of the modeling to the input variables: why does the figure they pick have any bearing on what the actual figure is?

They could have created a scenario 3 where specificity was set to 50% That doesn't mean specificity was actually 50%, just "this is what the figures might look like if it was"

Your entire argument that the PCR specificity was *not* 99.9% or better is based around the fact the paper didn't explicitly (it did implicitly) define the specificity in one scenario.

Talking of ignoring: what is your response to the statistical fact that there is a 0.006% for a test with a specificity of less than 99.9% to give the observed results?

[Scott777]

so they used the "high specificity" value of 99.9% for both.

You can just ignore the actal quotes I posted, and carry on making false claims in blissful ignorance, but it only exposes yourself.  Or, just admit you are wrong.  You had 2 weeks to think about it.

"we have estimated what prevalence would be in two scenarios using different possible test sensitivity and specificity rates."

[BeElBeeBub]

Well I'm away for 2 weeks, so you've got plenty of time to explain what the specificity is for scenario 2, or why it isn't given.  Good luck.  😁

I don't have to explain the specificity of scenario 2.  You're the one who doesn't understand what the scenarios represent.

They are sensativity (to input) studies. They are quite common in data analysis.  You take your data and assume some input variable (say selectivity) is as low as it could plausibly be, maybe even a bit lower. You run the numbers and see what pops out and how it differs with your actual estimate (in this case prevalence).

They explicitly say they are only looking at the specificity of the tests.

We have used Bayesian analysis to calculate what prevalence would be in two different scenarios, one with medium test sensitivity and the other with low test sensitivity.

Note there is no mention of specificity.  Of course you have to have a figure for specificity to run the calculations so they used the "high specificity" value of 99.9% for both.

They are explicit in the text that the specificity of the PCR tests is near 100%

Maybe use some of it to think how it's possible to run 207,730 PCR tests and only get 159 positive results is the chances of a false positive is the 1 in 100 you claim.

There is a 0 chance that a test with a FPR of 1 in 100 (99%) will give those results.

There's only a 0.006% chance a test with a FPR of 1 in 1,000 (99.9%) will give you those results

And the above assumes that none of the 159 positive results were true positives.

If only 59 of those 159 were true positives, your likely FPR starts heading past 1 in 175,000 (99.9994%)

Have a nice time! Hope the weather is kind

[Scott777]

When they set up scenario 1.they say they use "High specificity" and define that as 99.9%. for scenario 2 they also say they used "high specificity" but don't define it directly, presumably because they assume the reader will assume they mean the same. 

Well I'm away for 2 weeks, so you've got plenty of time to explain what the specificity is for scenario 2, or why it isn't given.  Good luck.  😁

[Scott777]

When they set up scenario 1.they say they use "High specificity" and define that as 99.9%. for scenario 2 they also say they used "high specificity" but don't define it directly, presumably because they assume the reader will assume they mean the same. 

But they don't mean the same, and it says so:
"Our data and related studies provide an indication of what these are likely to be. To understand the potential impact, we have estimated what prevalence would be in two scenarios using different possible test sensitivity and specificity rates."
Proving that your assumption of 99.9% for both scenarios is FALSE.  Now you'll need to explain what the specificity is for scenario 2, if it isn't what I said.  And they don't define "High specificity" as 99.9%.  That's nonsense.  99.9% could be called high, but "high" does not = 99.9%, and the above quote proves it isn't.

Yet despite that, you want to believe the figures apply to a completely different test that isn't mentioned in the paragraph. You want us to believe that when they say "this assay" they don't mean any of the assays in the study they just mentioned they mean a completely different assay they last mentioned in a previous paragraph

No, the whole study is about a PCR test, and section (5) is also about a single test and a single assay, however, other tests and assays are referred to (serology), but are not the main subject of the study.

[BeElBeeBub]

Here are the problems with your interpretation:

If you were correct, using correct grammar, the article SHOULD say "Evaluation of the test sensitivity and specificity of five immunoassays for SARS-CoV-2 serology has shown that the ELISA assay, as included in our study, has sensitivity and specificity...".  That way, we know exactly what they are referring to.  In the actual sentence, the Elisa assay is within a clause, and so not the main subject of the sentence.

"In both scenarios they use "high specificity", defined as 99.9% in scenario 1."  This is nonsense, because the 99.9% would not be contained ONLY within scenario 1 if it applied to both scenarios.

As it stands, you are suggesting NO specificity has been given for scenario 2, and that the specificity in scenario 1 is just a single number, without any range of variability, or margin of error.  Despite the five immunoassays ALL showing a range of variability, AND this "To allow for the fact that individuals are self-swabbing...  provides a lower bound."  So scenario 2 is explicitly intended to provide a lower bound, and yet you claim there is just a single number for the specificity.

Finally, "We have used Bayesian analysis to calculate what prevalence would be in two different scenarios".  Do you understand, if you apply the exact same specifity to a quantity of positive test results, then you get the same number of false positives, and therefore the same number of true positives, then the prevalence would be the same, contradicting the purpose of the 2 scenarios.

So, pray tell, why is the specificity not stated for scenario 2, why is the specificity without any range of error, and why would both scenarios have the same specificity if they are for measuring the possible range of prevalence?

The sentence is clear. They point to a study of 5 blood antibody assays, including the one they used for the study and give some performance figures for it.  Those figure exactly match the figures given for the assay they used (ELISA) in the study. 

They then say that thes figures given are better than the next best assay in the study and give the figures for that assay which also.exactly match the figures for the second best performing assay in the study.

Yet despite that, you want to believe the figures apply to a completely different test that isn't mentioned in the paragraph. You want us to believe that when they say "this assay" they don't mean any of the assays in the study they just mentioned they mean a completely different assay they last mentioned in a previous paragraph


As for the specificity in the 2.scenarios, the scenarios are just that. they are not the data, they are thought experiments about "what if".

When they set up scenario 1.they say they use "High specificity" and define that as 99.9%. for scenario 2 they also say they used "high specificity" but don't define it directly, presumably because they assume the reader will assume they mean the same.  

You don't use a range within a scenario because the point of scenarios is to do limit testing at a.single value. It is not unusual at all to only change one variable between scenarios

[BeElBeeBub]

Not correct.  You would expect, as an estimate, to see 2000 positive results.  It doesn't mean that's what you would get.  It just means you have to subtract 2000 from the actual result, because they cannot be taken at face value, so none of the 159 positives can be taken as true positives.

Oh my god,  you really are intent on proving how little you understand about probability and statistics! You are just talking innumerate gibberish.

The study says the FPR was very low (the specificity was high)

They point to the very low number of positives (159) from the 200k summer tests as proof.

You are grasping at grammatical straws

If the FPR was as high as you claimed you would expect to see around 2,000 positives in that test series. The chances of the FPR being as high as you claim and only getting 159 positives are astronomically small.

There is zero chance your theory that the FPR is around 1% is true. You have more chance.of winning the national.lottery every month for a year