Clearly, testing is a valuable tool and I'm sure everyone can see that, but it does generate positive results where previously they wouldn't have been recorded, simply because in many cases the infected person would have simply got better in a few days and never known that they had CV19.
What I can't understand is why the recipients of drive through testing aren't analysed separately, because including them leads to yet another change, which is at least the 3rd, in the way the figures are presented for public consumption. It is simply impossible to make valid judgements, decisions or plans if the data collection sources are repeatedly changed.
Obviously the scientists know this must be happening so why do they subscribe to presenting inaccurate figures?
Great point and not something I can personally answer with any real sense of accuracy, but in a generalist point of view, some data is better than no data, well, up to a point.
Going off on a bit of a tangent, but hopefully helps clarify what I’m thinking...
Lets say for example we wanted to look at the data set on empathy. Social science can do that. So we take 100 people and we’d expect to find a normal distribution in this sample population.
so this means about 34 of the folks are neither particularly empathic nor indifferent, about 25 will be fairly empathic, while another 25 will be quite indifferent. The 2 groups of 8 people bookending will be highly empathic and very indifferent. Ok, so this in my illustration shows a model of “some data”. It’s not particularly robust and quite subjective and tells us nothing of any subsets.
Now if we determine this group of 100 is made up of men and women, we would see two plots each showing a roughly normal distribution. We would likely see that 17 women compared within their subset also show neither empathy nor indifference. The men would also show this same pattern of distribution, but compare these now two independent sunsets and we would see a disparity between the two. It is likely that the 4 most indifferent women would compare well with the 4 most empathic men. The 4 least empathic men would be quite cold whilst the 4 most empathic women would be incredibly warm-hearted. This additional data now informs us better and we can use this to make better choices but the total absence of data leaves us totally uninformed.
So yes, the PCR amplification process in the way it’s being conducted (and I believe not every lab does exactly the same things) may well not be ideal, but for a rapid and low cost test it gives us a reasonable answer, that if it tells someone to stay home and their fine, no harm to health done. Possibly mental health, but then if 14 days isolation is tipping you over the edge, it would be my guess that you’ve got a lot of other complications going on that are more likely to be the root cause of your problems. If the test result is a false negative, sure that’s going to be more damaging.
If you look at it like this, a test that gives 100 % false positives isn’t such a bad thing. Not great, but not bad. A test that gave even just 50% negatives would be pretty much useless. But at anything below the 30% false negative region has a good bias. Now the question of is that good value, well it depends on how you frame the value proposition. On money alone, probably not. On lives alone... it makes a strong argument.
Furthermore, you also need to check how the process efficacy is being determined. You can’t say someone testing a voltage with a Ohm meter is wrong because your guessing it’s around the 240 volts mark. To test your hypothesis you need a gauge more robust than your experiment.
In summary, what we have is far from perfect, but it’s better than nowt
I would also add that what is presented to we the proletariat is almost always a dumbed down version of what the scientists will be using themselves and the media do have a propensity to fudge even this, particularly if they have their own agenda
