Meet the Other Phone. Only the apps you allow.

Meet the Other Phone.
Only the apps you allow.

Buy now

Please or to access all these features

Covid

Mumsnet doesn't verify the qualifications of users. If you have medical concerns, please consult a healthcare professional.

So they are openly saying they are counting tests twice?

59 replies

Coldwinds · 22/10/2020 09:10

This is why people question. Not because of disbelief of a virus but they don’t trust the data.

If you test positive it should only be ever counted as one test in one set period.

WTF are they playing at? How much money have these guys had?

OP posts:
frumpety · 22/10/2020 21:14

So if you had three tests in the period of Friday to Tuesday then it would only count as one test ?

Lougle · 22/10/2020 21:26

@frumpety

So if you had three tests in the period of Friday to Tuesday then it would only count as one test ?
Yes. 3 negative tests would result in 1 negative being recorded. 3 positive tests would result in 1 positive being recorded. Any combination of negative and positive tests would result in 1 positive being recorded.
frumpety · 22/10/2020 21:30

Thanks Lougle , forntunately they were of the negative variety Smile

Turtleshelly · 23/10/2020 02:06

The number of tests is overcounted NOT the number of results.

So 200,000 tests “processed” isn’t 200,000 people. It’s been found to be about 60-70,000 people.

But every positive result is counted once.

They’re exaggerating how many tests they do not the number who test positive. (And then the Mps lie and quote the figure as people tested but that’s another issue)

namechangefail2020 · 23/10/2020 05:43

Bloody hell it's a good job you lot are not the ones responsible for the data! I'm not even that bright but it's really obvious what is being said!! Stick to your day job!

Coldwinds · 23/10/2020 06:41

@namechangefail2020

Bloody hell it's a good job you lot are not the ones responsible for the data! I'm not even that bright but it's really obvious what is being said!! Stick to your day job!
Wow so early for arse hole comments...
OP posts:
namechangefail2020 · 23/10/2020 06:54

And defensive ones.....

MRex · 23/10/2020 07:17

It's been explained clearly by @Hardbackwriter for those who were unsure. It really doesn't make sense to keep leaving out tests for anyone who was first tested back in April, July etc because it gives a misleading reflection of both positivity and testing, there simply isn't a perfect answer. If you aren't used to reading data methodology information, then it's understandable it would take a moment to process, but then why have you suddenly started trying it now with this one statement?

CuriousaboutSamphire · 23/10/2020 08:56

[quote frumpety]@CuriousaboutSamphire the tests were friday to thursday , all negative.[/quote]
So, assuming that a test was given on the first Friday and last Thursday (I have imagined Wednesday as a non test day, one a day would be 6 tests). But even if there were 6 tests the answer would remain the same:

1 test in each of 2 weeks in both types of week! See below capitals = test day

th F S S M T w - Reporting Week 1, all one test as the result was the same each test

TH - Reporting Week 2

CuriousaboutSamphire · 23/10/2020 08:58

Toads. Maths let me down. Make Sunday and Wednesday non testing days.

Sorry Frumpety I was concentrating on making the reporting week boundary clear and forgot to adjust to your 5 tests over 7 days.

Coldwinds · 23/10/2020 09:18

@MRex

It's been explained clearly by *@Hardbackwriter* for those who were unsure. It really doesn't make sense to keep leaving out tests for anyone who was first tested back in April, July etc because it gives a misleading reflection of both positivity and testing, there simply isn't a perfect answer. If you aren't used to reading data methodology information, then it's understandable it would take a moment to process, but then why have you suddenly started trying it now with this one statement?
I’m not sure what you mean? Am I not supposed to try and understand the methodology of a data published on Public site because I’m not a data analysis Confused

Why is is it irritating you to the point your having to make a snipy comment? Why do you think this is the first time I’ve read data? I’m not the only one by the way who this this was potentially confusing.

So strange that I post something that actually quite a lot of people have found confusing yet I’ve been called stupid and treated with suspicion.. weird.

OP posts:
CuriousaboutSamphire · 23/10/2020 09:33

I’m not the only one by the way who this this was potentially confusing. Nope. And there wll be continued confusion for ages yet.

There is a great gap between the level of information that the public, media etc demand and the ability of a sizeable % of both to understand what that data means. Your isn't the only post that rails about inconsistent data. There's a whole set of threads that attempt to simplify it, it's been gping on since the beginnibgn of lockdown.

One point is that the data changes, quite regularly, depending on a whole range of factors, including sher amount of data, better data collection, understanding of the disease etc etc. Some see those changes as inconsistency others as necessary adjustments.

Another is the multiple data sets. As here with the different 'weeks'. They are necessary due to things we often have no knowledge of. The reporting week will be to accommodate weekends, some other data collection points etc etc. to the analysyst it doesn't matter. They deal with the data set they work with. To onlookers it seems odd and causes confusion.

But I keep coming back to one point: no matter how obvious Joe Public may think a perceived flaw is there are people whose whole lives is spent using that data. They will know far more than Joe does. Yet SM listens to Joe. Why is that?

Instead of getting annoyed that it doesn't make sense to you, sniping back at posters who snipe at you, you could try pen and paper. Like my attempt at trying to explain above. Sometimes going back to basics helps.

Basically, it doesn't matter if you/we don't understand it. There are plenty of experts around the world who do and the data is meant for them, not us. That we (in general) have recently found that such data has always been there, free to access, doesn't mean the data will change to make it more accessible to us. That isn't its purpose. You just have to accept that you either will or will not understand some/all of its complexities and move on!

RaspberryCoulis · 23/10/2020 09:37

The Scottish government has fudged this too and chopped and changed a lot of the ways they count positive figures.

blogs.gov.scot/statistics/2020/10/19/new-headline-measure-of-covid-19-test-positivity-rate/

They changed the way they were counting positive results and the percentage of people testing positive dropped from 17% to 6%. I also don't think it's fair to say that people need to educate themselves and get up to speed with statistics and epidemiology. That blog I linked to above is clearly written by a statistician and contains loads of technical jargon - change the focus to an alternative daily positivity metric - and so on.

There's no harm in having the science and stats there for people who want to read it. In fact, it's essential that they DO publish the numbers so people can crunch the figures and pull them up on decision making.

But most people aren't going to read statistics blogs and will just hear that the percentage positive has dropped from 17% to 6%. Statisticians and scientists may not be proficient in communication with the mass public. Information needs to be presented in simple, uncomplicated language, jargon-free. Or people will just switch off and not listen.

Bollss · 23/10/2020 09:41

But why so they need to count it if someone tests positive more than once? Because it's still only one case isn't it, it's just ongoing. By testing multiple times it gives the impression of more cases, more people affected.

Also that they'll count a positive test over a negative one, so if u test positive then negative in the same week they'll count you anyway even tho you haven't got it. Makes loadsa sense.

It's all bullshit isn't it!

CuriousaboutSamphire · 23/10/2020 09:50

No sweets, you just haven't understood what has been explained!

Then again I didn't quite understand your post either!

so if u test positive then negative in the same week they'll count you anyway even tho you haven't got it. ?????

Bollss · 23/10/2020 09:54

@CuriousaboutSamphire

No sweets, you just haven't understood what has been explained!

Then again I didn't quite understand your post either!

so if u test positive then negative in the same week they'll count you anyway even tho you haven't got it. ?????

It says with a positive test being prioritised over a negative test. Ie you have two tests in one week. One pos one neg. They're gonna count the pos test even tho the neg test if taken after has proved you do not have it.
Justforphoto · 23/10/2020 10:02

TrustTheGeneGenie

Yes because it is the number of new cases not the current number of positive cases, that's a different statistic again

Bollss · 23/10/2020 10:03

@Justforphoto

TrustTheGeneGenie

Yes because it is the number of new cases not the current number of positive cases, that's a different statistic again

But if they've had a negative test it's not a new case is it?
Justforphoto · 23/10/2020 10:08

Yes it is still a new case, even if they have since tested negative they have still had covid so still need to be recorded as a case.

Bollss · 23/10/2020 10:17

@Justforphoto

Yes it is still a new case, even if they have since tested negative they have still had covid so still need to be recorded as a case.
Or they've just had a false positive test ?

You cannot categorically say that you HAVE had covid if you test positive then within the same week negative.

CuriousaboutSamphire · 23/10/2020 10:36

FFS!

It's not infallible, no tests are! How do YOU suggest the issue is sorted? Write up with a better system, in detail, covering all possible factors, cohorts and confounding factors. Or is this the best you have? To cavil about something you don't really understand!

Bear in mind you are demanding that science prove a negative... quite literally!

Bollss · 23/10/2020 10:37

@CuriousaboutSamphire

FFS!

It's not infallible, no tests are! How do YOU suggest the issue is sorted? Write up with a better system, in detail, covering all possible factors, cohorts and confounding factors. Or is this the best you have? To cavil about something you don't really understand!

Bear in mind you are demanding that science prove a negative... quite literally!

I'm saying don't count a positive test as a case if a negative test directly follows it? That person won't be isolating so imo they're not a case.

You don't have to start being horrible about it just because I disagree that it's a case.

Jesus Christ.

Justforphoto · 23/10/2020 10:37

It's just a likely to be a false negative as a false positive. Truth is that they are still missing a lot of cases you only need to look at the ONS data. I know people are trying to claim that the numbers of positive cases are being exaggerated but they aren't.

Bollss · 23/10/2020 10:40

@Justforphoto

It's just a likely to be a false negative as a false positive. Truth is that they are still missing a lot of cases you only need to look at the ONS data. I know people are trying to claim that the numbers of positive cases are being exaggerated but they aren't.
Surely that proves that it's not all as horrific as you might think? If we have thousands more cases than we know about but still deaths are low that's a good thing no?

I still disagree that someone who has tested negative after testing positive should be counted as a case.

CuriousaboutSamphire · 23/10/2020 10:47

I'm saying don't count a positive test as a case if a negative test directly follows it? That person won't be isolating so imo they're not a case. Your opinion? OK! Call the WHO!

You don't have to start being horrible about it just because I disagree that it's a case. I'm not being horrible I'm frustrated. You don't understand. You have an opinion. You are not listening to the opinion of others. You are dismissing the combined decisions of data analysts the world over.

Jesus Christ. Indeed!