General Discussion

Ms. Toad

(38,824 posts)

30. You are the one asserting a number that cannot be supported

Mon Jul 5, 2021, 09:43 PM

Jul 2021

unless the sample is representative.

Its on you to justify that the sample is not only large enough, but representative.

It is math illiteracy, at its most basic, to pretend that you can just multiply the ratio of positive tests in a self-selected sample times the population at large. And since I have two math degrees, and taught math for 11 years, I'm well-suited to recognize math illiteracy. Identifying it as such has nothing to do with confirmation bias. I follow the math where it leads me. Not vice versa.

The most basic question you have to ask yourself when trying to justify extrapolating from a very small sample to a population as a whole, is to identify whether the populations are similar enough to justify it.

You have not justified why your sample, whether 10,000 or 7 million is representative of the population of the UK as a whole as to infection with COVID 19.

Edit history

Please sign in to view edit histories.

Recommendations

0 members have recommended this reply (displayed in chronological order):

55 replies

= new reply since forum marked as read

Highlight:

The UK currently has 1,657,270 positive cases of COVID-19 [View all] Yavin4 Jul 2021 OP

Correct me if I'm wrong, but they're not testing random samples of the population, are they? RockRaven Jul 2021 #1

They're doing over 7 million tests a week Yavin4 Jul 2021 #3

The large size of the sample doesn't fix the selection bias problem. RockRaven Jul 2021 #5

I don't have insight into who or how their tests are done. Yavin4 Jul 2021 #8

This message was self-deleted by its author BannonsLiver Jul 2021 #9

The opposite, actually. RockRaven Jul 2021 #13

No, I had it just right. BannonsLiver Jul 2021 #16

Not when the large sample is selected in a non-random manner. The size does not make it RockRaven Jul 2021 #12

I understand sampling bias. Yavin4 Jul 2021 #18

Obviously, you don't. Ms. Toad Jul 2021 #31

The UK has conducted over 200 million tests. Yavin4 Jul 2021 #34

You are still missing the difference between size of sample, Ms. Toad Jul 2021 #37

And that is precisely why you can't extrapolate. Ms. Toad Jul 2021 #20

I have been tracking Russia and the UK BigmanPigman Jul 2021 #2

Yep. Yavin4 Jul 2021 #4

Except Sputnik V... nt Shermann Jul 2021 #11

What about the US? ecstatic Jul 2021 #6

People generally test when there is a reason to test. Ms. Toad Jul 2021 #7

7 million people? Yavin4 Jul 2021 #10

A representative sample must still proportionately mirror the population. Ms. Toad Jul 2021 #17

It's like extrapolating STD positivity rates to the general population NickB79 Jul 2021 #19

Perfect example. Ms. Toad Jul 2021 #22

You are correct, but it's not the same as extrapolating STD positivity rates. Yavin4 Jul 2021 #29

They test over 1 million people a day. Yavin4 Jul 2021 #21

Quantity means next to NOTHING Ms. Toad Jul 2021 #25

You are making assumptions without any evidence or facts. Yavin4 Jul 2021 #28

You are the one asserting a number that cannot be supported Ms. Toad Jul 2021 #30

When you talk about 7 million and not 700 Tribetime Jul 2021 #32

Quantity only alters the need for representation Ms. Toad Jul 2021 #33

The UK leads all of Europe in the number of tests done. Yavin4 Jul 2021 #35

Completely irrelevant - Ms. Toad Jul 2021 #38

Numerous people in this thread simply do not understand basic statistics. Yavin4 Jul 2021 #43

No; schools do a lot of regular tests, and so do care homes and hospitals muriel_volestrangler Jul 2021 #45

Schools, care homes, people attending Wimbeldon, people traveling abroad, people coming home Yavin4 Jul 2021 #48

Wrong. I think you are now winding us up, but it isn't funny. muriel_volestrangler Jul 2021 #54

You are the one who does not understand basic statistics, Ms. Toad Jul 2021 #46

You need to show me where the bias is in the sample. Yavin4 Jul 2021 #47

A sample is presumed to be biased, Ms. Toad Jul 2021 #50

There's a historic example for this BGBD Jul 2021 #36

Yup. But that point isn't getting across, for some reason. n/t Ms. Toad Jul 2021 #39

Election sampling is completely different Yavin4 Jul 2021 #42

Then explain the mix - Ms. Toad Jul 2021 #51

In the well vaccinated Seattle area, cases are dropping ismnotwasm Jul 2021 #14

Probably already said Dave says Jul 2021 #15

Many sick are getting tested IbogaProject Jul 2021 #23

People in the UK get tested for a variety of reasons. Yavin4 Jul 2021 #26

At that rate, the entire nation would be infected in 40 days NickB79 Jul 2021 #24

Not quite. Yavin4 Jul 2021 #27

From the Office for National Statistics, it's about 257,000 at 26th June muriel_volestrangler Jul 2021 #40

Please show me evidence that the 7 million tests per week were NOT at random. Yavin4 Jul 2021 #41

Sure, though everyone has already explained that tests are mostly done for a purpose muriel_volestrangler Jul 2021 #44

Again, I'm using your governments official daily reports which when they support the groupthink here Yavin4 Jul 2021 #49

The point is that you are making an unjustified assumption that all the people tested muriel_volestrangler Jul 2021 #53

I got pretty darn close just estimating the testing mix and estimated positivity within the groups Ms. Toad Jul 2021 #52

Fine. I will use your estimate of 600K positive cases Yavin4 Jul 2021 #55