First of all, hello and congratulations to everybody for creating a bubble where you can protect yourself from criticism of your ideas and safely keep believing what you want to believe, free of interference from people who might persuade you otherwise.
Sorry to burst your bubble, but the government is aware of the issue of false positives, which is why they have told labs to run twice tests that give a weak result:
Besides, the cases graph has been going up with an increasing slope, which strongly indicates that it's growing exponentially. This doesn't happen merely because of changes in testing or the number of people getting tests, since people getting tests aren't increasing exponentially. The government also checks the positivity rate, ie, how many tests come out positive out of all the tests done. The positivity rate has increases, which again shows that there is a second wave.
You are entirely correct that the graphs are misleading in that in the first wave there was limited testing. The BBC is lately indicating the difference in testing regime with a line.
I'm sure many of you are in businesses that are badly affected by the restriction measures, but spreading your belief that the measures are unwarranted, which you mostly believe because you want to believe it, won't help you at all. If people pay attention to your theories, then the cases will spread further, and there will be more restrictions needed. If you want to reopen pubs and other businesses quickly, the best attitude is to support the restrictions so that they will be lifted as quickly as possible.
By the way, the UK government statisticians are excellent. I say it as somebody with a maths degree. So don't try to convince yourselves that you are out-smarting them, because you really aren't. However, if this piques your interest in statistics, congratulations. I recommend that you post all your thoughts and questions to a forum with actual statisticians, instead of staying in this bubble.
First of all, hello and congratulations to everybody for creating a bubble where you can protect yourself from criticism of your ideas and safely keep believing what you want to believe, free of interference from people who might persuade you otherwise.
Sorry to burst your bubble, but the government is aware of the issue of false positives, which is why they have told labs to run twice tests that give a weak result:
Besides, the cases graph has been going up with an increasing slope, which strongly indicates that it's growing exponentially. This doesn't happen merely because of changes in testing or the number of people getting tests, since people getting tests aren't increasing exponentially. The government also checks the positivity rate, ie, how many tests come out positive out of all the tests done. The positivity rate has increases, which again shows that there is a second wave.
You are entirely correct that the graphs are misleading in that in the first wave there was limited testing. The BBC is lately indicating the difference in testing regime with a line.
I'm sure many of you are in businesses that are badly affected by the restriction measures, but spreading your belief that the measures are unwarranted, which you mostly believe because you want to believe it, won't help you at all. If people pay attention to your theories, then the cases will spread further, and there will be more restrictions needed. If you want to reopen pubs and other businesses quickly, the best attitude is to support the restrictions so that they will be lifted as quickly as possible.
By the way, the UK government statisticians are excellent. I say it as somebody with a maths degree. So don't try to convince yourselves that you are out-smarting them, because you really aren't. However, if this piques your interest in statistics, congratulations. I recommend that you post all your thoughts and questions to a forum with actual statisticians, instead of staying in this bubble.
1. I agree with you on the bubble thing.
2. I very strongly disagree with you what you saying about "restrictions". Restrictions because of what ?! Pergaps you can enlighten me, because I see absolutely no reason for any safety measures or lockdown or even any restrictions. And if you thinking about covid-19 as an answet to this, please attach evidence that it exists as I was unable to find any since the day I started looking for it, which was 16th of April 2020.
3. Because you are refering to some kind of forum where "the real" statisticians dwell, perhaps you can share the link to it and just a little add-on to this link I would like to ask you if you can find on that particular forum you refering to a statistical calculation which supports the lockdown, because there is about 540 people dying almost every day of different cancers, that weren't treated because hospital admissions are restricted to covid patients. So maybe there will be statistical explanation that the lockdown and restrictions actually work.
Thanks
Sorry to burst your bubble, but the government is aware of the issue of false positives, which is why they have told labs to run twice tests that give a weak result:
That will be good. Let's see how many they do.
Besides, the cases graph has been going up with an increasing slope, which strongly indicates that it's growing exponentially. This doesn't happen merely because of changes in testing or the number of people getting tests, since people getting tests aren't increasing exponentially.
Please look again. There was a step jump in %cases/tests on 6th September. Please provide a chart that shows an exponential increase - even including false positives.
By the way, the UK government statisticians are excellent. I say it as somebody with a maths degree. So don't try to convince yourselves that you are out-smarting them, because you really aren't.
There we agree - but I am not trying to outsmart them - it is the misuse of their data that is wrong.
Congratulations on you maths degree. I read maths at Trinity College, Oxford - although I eventually got an engineering degree and I am a retired aircraft stress engineer.
If there is anyone uncertain about the implications of the new September “protocols” on reporting results , Nicola Sturgeon clearly wasn’t .
From 9th of September
NICOLA Sturgeon said it was better to "err on the side of caution" on Covid testing even if it means generating more false positives.
It comes after Public Health England issued new guidance to laboratories south of the border asking them to set a threshold for the amount of viral material needed before a positive test would be considered accurate.
https://www.heraldscotland.com/news/18706347.sturgeon-says-rather-catch-many-covid-cases-possible---even-means-false-positives/
On the next bit I would really appreciate input from any other analytical chemists, engineers and especially biotech people; because this DL, instrument detection limit (IDL), the method detection limit (MDL), the practical quantification limit (PQL Limit of Blank (LoB), Limit of Detection (LoD), and Limit of Quantitation (LoQ) has always made my headache and I try and forget it all as soon as possible.
As I understand in my simple way you determine the mean and standard deviation on a set of blank samples.
Select a confidence limit that you are happy with eg the false positive rate.
And from that determine or derive your limit of detection.
Although I suppose you can go the other way and pluck your LOD out of thin air and determine you false positive rate.
Sticking to the conditions of the situation we are dealing with.
Or to put it another way if they have or “know” LOD then the have and “know” the false positive.
OK so what does that mean;in this case?
They take a sample with no viral RNA in it and analyse it 50 times or take 50 samples with no viral RNA in them.
As it goes I think in this kind of analysis, some of the samples will start to glow in the dark or whatever at 39 cycles others at 42, 45 , 51 etc etc.
The average of the 50 samples might be 45 and the standard deviation 5.
Standard deviation is a measure of how spread out values from the average; it is a little bit like the range.
For instance normally
90% of the values fall within a range of 1.6 standard deviations ; 45 +/- 8
95% of the values fall within a range of 2 standard deviations ; 45 +/-10
99% of the values fall within a range of 2.6 standard deviations ; 45 +/-13
We will take that last one 45 +/-13; so we would expect 99 times out of 100 to get a “result” on a sample with no RNA in it at between 32 and 58 cycles.
We have just determined that 1% of blanks either give a result above 58 cycles or below 32.
So 0.5% of samples with no viral RNA will give a “result” at or below 32 cycles [ false positive].
We are not bothered here about ‘results’ at over 58 cycles.
So in other words if we set the limit of detection [essentially the level at which we say it is there or report it as a positive] at 32 then we will only be [false positive] wrong 0.5% of the time.
That might not be good enough if you are worried about the consequences of nuclear meltdown or warp core breach in which case you might want to ramp it up to one in a million and +/- 10 standard deviations or whatever.
As it is popular to do CV’s ; I am a investigative analytical chemist of 35 years
experience; a bit different to ploddy routine analysis.
I have published several papers on novel analytical methods; with shoddy and lamentable attention to LOD, LOQ, repeatability data, but at least it was always there.
Can do a bit of microbiology if I have to
And of course I am a anti authoritarian revolutionary Marxist . libertarian communist, thus a historical and not a popular materialist and thus I don’t believe this ultimately about stupid or crazy people.
It has to be about the economic interests of ruling class; somehow.
I have not read recent posts on this thread yet just had a late and horrid day on false positive patulin results.
By the way, the reference to a second test came from the Telegraph, dated 8th September. The actual government document, previously referred to in this thread, is here:
https://www.gov.uk/government/publications/sars-cov-2-rna-testing-assurance-of-positive-results-during-periods-of-low-prevalence/assurance-of-sars-cov-2-rna-positive-results-during-periods-of-low-prevalence
Concerning positive test results at the limits of detection, they say
These results should not be labelled as false positive in the absence of complete case-by-case work-up including retesting of existing samples, resampling for new testing and determination of clear evidence of analytical errors or laboratory contamination.
Then, in Recommended Actions, they say:
All laboratories should determine the threshold for a positive result at the limit of detection based on the in-use assay.
It is necessary to strike a balance between the risk of false positive test results and an acceptable level of delay in test turnaround time (time taken to report results).
They then go on to explain what to do, including a second test, if the chosen threshhold is reached. This is as long as a piece of string.
A further point (I may be wrong here, so please correct me if necessary), is that false positives occur due to other reasons and not just at these limits.
The point that I am emphasising in this thread is that there has been a jump in the cases/tests that occurred at the same time as the test rate was ramped up 3x. It needs explaining.
