England to surpass Wales again

[UncleEbenezer]

The only thing approaching a proper controlled experiment would be two equal populations: one masked, one not.

Along the lines of this ?

https://www.cam.ac.uk/research/news/upg ... amatically

Ah yes. Hospital staff.

If you're extending that to the general population, do you think by analogy that all drivers should wear all the protective gear racing drivers use?

[servodude]

- but it's a really long bow to take those and extrapolate to "masks increasing transmission"
- sd

Of course! I was positing possible explanations. The evidence (insofar as any exists) is in the statistics, not in my observations nor in that picture.

We know covid incidences rose (from a low level) when masks became mandatory 16 months ago. I publicly predicted that.

This time, we have something closer to a control sample, with Wales having had the mask mandate all along. I'm offering another prediction. Let's see where we are in a month's time.

I think there's a huge risk of correlation/cause conflation thing going on if population data are considered in isolation (particularly with respect to projection on to a single variable)
- we have all seen how the data shows "lockdowns" are always be called just before the virus would have burned itself out anyway
- and how ghostbusters is deadly
- the truth is this is a huge and multi-dimensional problem with massively varying gains in the different variables/dimensions

Properly worn masks serve as filters - they do that - we know they do - in both directions
We know how filters work
- we know the type of filters that will help with SARS-COV2 (masks that don't aren't recommended)
wearing masks indoors properly reduces the amount of virus floating about in the air
- that's the same reason ventilation is recommended (that is "a good thing")
- and the same reason we isolate and filter the return air from a covid patient on an NIV

I have no doubt that the proportion of crappy mask etiquette makes mask wearing mandates/recommendations less effective than they could be (especially at the personal level for the likes of those that breathe in through the unhindered nose and out through their mouth )
- but to make the situation worse (from a transmission perspective) overall?? - can't see it
- certainly can make it feel worse - suffering from stupid levels of hayfever the humidity caused by masks makes everything a lot more sensitive

- sd

[Lootman]

When I read the OP there weren’t yet any replies. I didn’t reply because I couldn’t understand what was meant - I thought it was a form of sarcasm. I’m really surprised that you actually seem to believe that masks promote contagion UE. Not at all what I’d have expected of you.

Uncle's thought experiment is in fact far more in keeping with true scientific method than those who merely reject his idea out of hand because it is not what the government keeps telling us. Science advances not by mindlessly trotting out existing theories but rather by challenging them. Uncle advances a counter-theory but he also defines a real world example where the resultant statistics might reveal something new and interesting. Everyone assumes that masks are always good in every case, but are they?

For those with the time I would recommend "The Logic of Scientific Discovery" by Karl Popper. It's about 60 years old now but I am not aware of a later book that better describes how science advances. Here is a teaser: Science does not advance by armchair experts blindly accepting what the supposedly real experts keep telling us. After all the government may have many motives to get us to believe certain things and hates those who question them. But science does advance through what Popper describes as falsifiability - the relentless process of questioning current theories and looking for real-life examples that contradict what the existing science tells us. This eventually leads to old theories being replaced by new theories that better fit reality.

So what if Uncle is right here and the England/Wales stats continue to show that infection rates correlate to mask adoption? Will the cynics ignore any data that doesn't fit their theory? Or will they adjust their theory to better fit the data? It seems to me that there are a few here are a little too vested in the comfort they derive from believing everything they are told.

I admit I don't know for sure whether masks can actually promote infection. I can see how they could however. But what is more important is to retain an open mind and be willing to consider data that might critically question what we are being told. It seems at least possible that the population walking around with pieces of cloth attached to their faces which are laden with virus for hours at a time might just possibly have some results that we do not like. A true scientist allows for that possibility and investigates it. A true scientist does not mock anyone who seeks to falsify an existing theory, because science does not fear having its theories tested. The insecure might, however.

[servodude]

The only thing approaching a proper controlled experiment would be two equal populations: one masked, one not.

Along the lines of this ?

https://www.cam.ac.uk/research/news/upg ... amatically

Ah yes. Hospital staff.

If you're extending that to the general population, do you think by analogy that all drivers should wear all the protective gear racing drivers use?

Are you saying masks work too well?

Saying that I do remember figures that concluded cycling helmets would be of more benefit to pedestrians than cyclists
It was due to the number and type of injuries sustained and the travelling speed (which dwarfed every other variable)

- sd

[UncleEbenezer]

I think there's a huge risk of correlation/cause conflation thing going on if population data are considered in isolation (particularly with respect to projection on to a single variable)

Absolutely. But I'm in no position to conduct a better-controlled experiment, and those who could have avoided doing so,

Best I can do is keep predicting whenever a significant change in the law happens. My record to date on mask laws is two out of two; this is a third.

[UncleEbenezer]

Saying that I do remember figures that concluded cycling helmets would be of more benefit to pedestrians than cyclists
It was due to the number and type of injuries sustained and the travelling speed (which dwarfed every other variable)

- sd

Cycling helmets is a subject filled with disinformation. The most widely-claimed figure comes from a study in Washington whose samples were helmeted white kids in a genteel suburb vs unhelmeted black kids in the ghetto. And when Victoria (Oz) introduced the first law, the effects were massively misrepresented (see the work of Dorothy Robinson for analysis).

[servodude]

I think there's a huge risk of correlation/cause conflation thing going on if population data are considered in isolation (particularly with respect to projection on to a single variable)

Absolutely. But I'm in no position to conduct a better-controlled experiment, and those who could have avoided doing so,

Best I can do is keep predicting whenever a significant change in the law happens. My record to date on mask laws is two out of two; this is a third.

Yeah but you can see how the effect of masks could be overwhelmed easily by concurrent events leading to a false conclusion?
Because I think in motors - I would consider masks to be just affecting the acceleration a bit whereas things like social distancing (or whatever the opposite is ?!) are shifting gears

Have you looked for any better-controlled "experiments' - or cleaner delineated policy changes?
There was one by the Burnett institute down this way;it was looking at the mask mandate in Vic introduced at them point where they had tried just about everything else in mid 2020
... give us a mo.

Right it's this one: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0253510

The conclusion was

The mandatory mask use policy substantially increased public use of masks and was associated with a significant decline in new COVID-19 cases after introduction of the policy. This study strongly supports the use of masks for controlling epidemics in the broader community.

- give it a read if you'd like to see how they got to that conclusion (it's not very controversial)

It appears the images are available to download so I'm going to link to this one



- basicaly the trend changes direction exactly where you would expect it to if masks were doing what they were meant to
- and you can see it in this data because there wasn't really a whole lot of anything else going on (trust me on that )
- sd

[Clariman]

The resistance to masks by a significant proportion of people in England completely baffles me. And it is all whipped up by the media harping on about it.

I'm English but live in Scotland. The masks have been a legal requirement here all along - and there has never been any significant fuss about it. It is simply accepted that we need to protect each other and ourselves and that masks help do that. It is not a major imposition. No big deal. Just get on with it.

Genuinely baffled
Clariman

[servodude]

Genuinely baffled

If that's not already a euphemism for wearing a legally compliant face covering it should be

[daveh]

The resistance to masks by a significant proportion of people in England completely baffles me. And it is all whipped up by the media harping on about it.

I'm English but live in Scotland. The masks have been a legal requirement here all along - and there has never been any significant fuss about it. It is simply accepted that we need to protect each other and ourselves and that masks help do that. It is not a major imposition. No big deal. Just get on with it.

Genuinely baffled
Clariman

I agree. I was in the supermarket last night. Big Tesco in Danestone Aberdeen. Not massively busy, but I'm pretty certain I saw more than 100 people and only 3 were not wearing a mask (and I guess at least one of those had an exemption as his partner was masked up). So the % wearing masks was well into the high 90s % wise. I've been at work (Uni research facility) since the end of first lockdown in July 2020 and we all mask up when not sat at our desks ( either single occupancy offices or well spaced multi person offices) and I don't find it particularly difficult or much of an imposition.

[redsturgeon]

The resistance to masks by a significant proportion of people in England completely baffles me. And it is all whipped up by the media harping on about it.

I'm English but live in Scotland. The masks have been a legal requirement here all along - and there has never been any significant fuss about it. It is simply accepted that we need to protect each other and ourselves and that masks help do that. It is not a major imposition. No big deal. Just get on with it.

Genuinely baffled
Clariman

Trying to avoid politics but when you know that the leader of the government and most of his ministers clearly do not think mask wearing is a good idea then what follows?

John

[servodude]

The resistance to masks by a significant proportion of people in England completely baffles me. And it is all whipped up by the media harping on about it.

I'm English but live in Scotland. The masks have been a legal requirement here all along - and there has never been any significant fuss about it. It is simply accepted that we need to protect each other and ourselves and that masks help do that. It is not a major imposition. No big deal. Just get on with it.

Genuinely baffled
Clariman

I agree. I was in the supermarket last night. Big Tesco in Danestone Aberdeen. Not massively busy, but I'm pretty certain I saw more than 100 people and only 3 were not wearing a mask (and I guess at least one of those had an exemption as his partner was masked up). So the % wearing masks was well into the high 90s % wise. I've been at work (Uni research facility) since the end of first lockdown in July 2020 and we all mask up when not sat at our desks ( either single occupancy offices or well spaced multi person offices) and I don't find it particularly difficult or much of an imposition.

I'm beginning to think that we've seen the reason for the apparent push back already in this thread.
There are a few facets but basically:
- they are an imposition
- they can be uncomfortable
- you can spin the data to look like they don't help

And the need to sell "whatever passes for news"

You can convince folk to be aggrieved about almost anything
- basically you do it by telling them obliquely that you've done their thinking for them and "oooh isn't it awful!"

Different papers sell better in different places and have different relationships with (or opinions about) who is in charge
- but really most folk aren't giving their "opinions" that much thought (they just have them) and the papers end up full of the extreme positions in a positive feedback loop (or downward spiral depending on your view)

So really I don't think that many people care either way; but those that do will massively sway the perception of how you perceive people feel if you look at "the media"

Ya ken? (that's for you Doric Loons in "furry boots" toon)

-sd

[XFool]

The only thing approaching a proper controlled experiment would be two equal* populations: one masked, one not.

From July to now we had something close to that. Wales with the mask rule, England without. Welsh infection rates soared above English ones.

But that's in the past and present. It's not a valid subject for a prediction posted now. So instead I'm predicting the outcome of the rules converging.

Surely the one thing you need for a realistic controlled experiment is control, so that you can be sure of constancy in the populations, that persists over the time of the trial, apart from the one thing you are varying.

Have we not recently seen an example of this going wrong? Recently circulating on the Internet - and turning up in all the usual places - was a chart derived (presumably honestly) from ONS data 'demonstrating' very clearly and visibly that the UK vaccinated population had a higher death rate than the unvaccinated population. The implicit conclusion being: "Don't get vaccinated".

And it was true! But...

I don't have statistical skills but, when I saw this, it rapidly occurred to me there was a problem. Bearing in mind the chart covered the year (2021) when the population was being vaccinated and the organisation of that was by age, starting with the oldest first. This meant the two populations, Vaccinated & Unvaccinated, were changing during the year with people passing from the U to the V population based on age! This meant the V population would inevitably have a higher mean age than the U population, and what increases risk of death more than age?

A proper professional analysis to the situation can be found here:

Simpson’s Paradox and Vaccines

https://covidactuaries.org/2021/11/22/simpsons-paradox-and-vaccines/

Stuart explains how a statistical quirk led to the appearance of higher mortality rates for some vaccinated people in a recent ONS release

It seems all too easy to be caught out by over simplistic interpretations, or seeing what you want to see.

[Watis]

The resistance to masks by a significant proportion of people in England completely baffles me. And it is all whipped up by the media harping on about it.

I'm English but live in Scotland. The masks have been a legal requirement here all along - and there has never been any significant fuss about it. It is simply accepted that we need to protect each other and ourselves and that masks help do that. It is not a major imposition. No big deal. Just get on with it.

Genuinely baffled
Clariman

I agree. I was in the supermarket last night. Big Tesco in Danestone Aberdeen. Not massively busy, but I'm pretty certain I saw more than 100 people and only 3 were not wearing a mask (and I guess at least one of those had an exemption as his partner was masked up). So the % wearing masks was well into the high 90s % wise. I've been at work (Uni research facility) since the end of first lockdown in July 2020 and we all mask up when not sat at our desks ( either single occupancy offices or well spaced multi person offices) and I don't find it particularly difficult or much of an imposition.

Just back from a trip to Tesco in England. Mask wearing back to almost 100% (two abstainers - who may be exempt anyway).

Prior to today, mask wearing had dropped steadily since July to about 60 - 75%

Watis

[Nimrod103]

The only thing approaching a proper controlled experiment would be two equal* populations: one masked, one not.

From July to now we had something close to that. Wales with the mask rule, England without. Welsh infection rates soared above English ones.

But that's in the past and present. It's not a valid subject for a prediction posted now. So instead I'm predicting the outcome of the rules converging.

Surely the one thing you need for a realistic controlled experiment is control, so that you can be sure of constancy in the populations, that persists over the time of the trial, apart from the one thing you are varying.

Have we not recently seen an example of this going wrong? Recently circulating on the Internet - and turning up in all the usual places - was a chart derived (presumably honestly) from ONS data 'demonstrating' very clearly and visibly that the UK vaccinated population had a higher death rate than the unvaccinated population. The implicit conclusion being: "Don't get vaccinated".

And it was true! But...

I don't have statistical skills but, when I saw this, it rapidly occurred to me there was a problem. Bearing in mind the chart covered the year (2021) when the population was being vaccinated and the organisation of that was by age, starting with the oldest first. This meant the two populations, Vaccinated & Unvaccinated, were changing during the year with people passing from the U to the V population based on age! This meant the V population would inevitably have a higher mean age than the U population, and what increases risk of death more than age?

A proper professional analysis to the situation can be found here:

Simpson’s Paradox and Vaccines

https://covidactuaries.org/2021/11/22/simpsons-paradox-and-vaccines/

Stuart explains how a statistical quirk led to the appearance of higher mortality rates for some vaccinated people in a recent ONS release

It seems all too easy to be caught out by over simplistic interpretations, or seeing what you want to see.

It was exactly the same ststistical analysis problem over 100 years ago, when they introduced tin helmets for British soldiers in the trenches, and found that they were treating far more head wounds. It took them a while to realise they were treating far more wounds because there were far fewer deaths.

I wish they would publish week by week deaths and hospitalizations broken down by age, sex and vaccination status. Sometimes this data is mentioned by individual specialists, but AFAIUI it is not routinely issued, presumably due to issues of privacy. I think it might provide much more of a stimulus to getting jabbed.

[servodude]

Simpson’s Paradox and Vaccines

https://covidactuaries.org/2021/11/22/s ... -vaccines/

Stuart explains how a statistical quirk led to the appearance of higher mortality rates for some vaccinated people in a recent ONS release

Yeah The Simpson's Paradox is a classic (and covered recently in the More or Less podcast) but I'm not sure it really fits UE's mask interpretation
- it's more of a "post ergo propter hoc" situation (with a bit of confirmation bias because of prediction?) wrapped up in morass of multivariate complication
- although I suppose the lack of nuance because of treating everything as a fungible mass could be a Simpson's style thing; I'm more used to it catching folk out by them focusing on the local effects

-sd

[XFool]

Simpson’s Paradox and Vaccines

https://covidactuaries.org/2021/11/22/s ... -vaccines/

Stuart explains how a statistical quirk led to the appearance of higher mortality rates for some vaccinated people in a recent ONS release

Yeah The Simpson's Paradox is a classic (and covered recently in the More or Less podcast) but I'm not sure it really fits UE's mask interpretation

No, I wasn't suggesting an exact analogy, rather it was an example of how these things are tricky and can catch one out. Especially so when really it is not a "controlled experiment".

My original opinion was that UE will be 'correct', in the sense that, if the new variety is very infectious, than of course cases must go up - masks or no masks. That's what you would expect.

i.e. His prediction will be consistent with the events and therefore won't be falsified by them, by being inconsistent with them*. But how does that prove it is actually true in these complicated circumstances? (Doesn't mean it isn't!)

I did hear some dude called Popper said something about this kind of thing. But that was 60 years ago...

* Unless it turns out it is!

[Lootman]

Karl Popper was a professor at LSE and one of the most esteemed British thinkers of the mid 20th century.

The really interesting thing about this topic is not the debate about the effectiveness of masks. But rather how a few here seem to feel genuinely threatened by a different idea. Uncle fired a cannon and the usual suspects immediately marched towards the fire, as if their very lives were under threat. Why does alternative thinking scare some so much?

Moderator Message:
Have edited out comments directed at another user, whose post has been deleted. Clariman

[UncleEbenezer]

My original opinion was that UE will be 'correct', in the sense that, if the new variety is very infectious, than of course cases must go up - masks or no masks. That's what you would expect.

You haven't even read what I've been posting this entire thread. Repeatedly. You just keep posting that irrelevant strawman that knocks down an argument I never made.

For example:

But in any case, I wasn't talking about a rise or fall in raw numbers. I was talking about the comparison between two samples, England and Wales. That's as near as we're going to get to a proper controlled experiment:

And for what it's worth, my education is in maths, and I've worked some of my professional career (admittedly not very much of it) as professional statistician. I'm not an epidemiologist, but I know a lot about samples, comparisons, and uncertainty. Hence for example my cautious language in engaging with excellent points made by servodude in this thread.

[XFool]

My original opinion was that UE will be 'correct', in the sense that, if the new variety is very infectious, than of course cases must go up - masks or no masks. That's what you would expect.

You haven't even read what I've been posting this entire thread. Repeatedly. You just keep posting that irrelevant strawman that knocks down an argument I never made.

Then could you please clarify, or recast, what you are saying. Because I am becoming increasingly confused as to what exactly you are claiming.

For example:

But in any case, I wasn't talking about a rise or fall in raw numbers. I was talking about the comparison between two samples, England and Wales. That's as near as we're going to get to a proper controlled experiment:

But my understanding(?) of the comparison was that you were predicting that the English population case rate was now going to rise to match the Wales case rate (due to recent mask rules in England). Is this so or not?

Seems [those in power] have declared masks mandatoy in England again. So now the scene is set for us to close the gap in infection rates with Wales, and resume the lead we consistently had before the [Deletion] rule was dropped in England in July.

My point was and is: Expect the English case rate to increase anyway (if variant really is more infectious). And no "controlled experiment".