by S. Stanley Young and Warren Kindzierski
Climate Etc. recently carried several insightful posts about How we fool ourselves. One of the posts – Part II: Scientific consensus building – was right on the money given our experience! The post pointed out that… ‘researcher degrees of freedom’… allows for researchers to extract statistical significance or other meaningful information out of almost any data set. Along similar lines, we offer some thoughts on how others try to fool us using statistics (aka how to lie with statistics); others being epidemiologists and government bureaucrats.
We have just completed a study for the National Association of Scholars  that took a deep dive looking at flawed statistical practices used in the field of environmental epidemiology. The study focused on air quality−health effect claims; more specifically PM2.5−health effect claims. However, the flawed practices apply to all aspects of risk factor−chronic disease research. The study also looked at how government bureaucrats use these claims to skew policy in favor of PM2.5 regulation and their own positions.
All that we discuss below is drawn from our study. Americans need to be aware that current statistical practices being used at the EPA for setting policy and regulations are flawed and obviously expensive. Viewers can download and read our study to decide the extent of the problem for themselves.
Unbeknownst to the public and far too many academic scientists, modern science suffers from an irreproducibility crisis in a wide range of disciplines—from medicine to social psychology. Far too frequently scientists are unable reproduce claims made in research.
Given the irreproducible science crisis, we completed a study for the National Association of Scholars (NAS) in New York as part of the Shifting Sands project. The project—Shifting Sands: Unsound Science and Unsafe Regulation—examines how irreproducible science negatively affects select areas of government policy and regulation in different federal agencies.
Our study investigated portions of research in the field of epidemiology used for US Environmental Protection Agency (EPA) regulation of PM2.5. This research claims that particulate matter smaller than 2.5 microns (PM2.5) in outdoor air is harmful to humans in many ways. But is the research on PM2.5 and the claims made in the research misleading?
2. Bias in academic research
Academic researcher incentives reward exciting research with new positive (significant association) claims—but not reproducible research. This encourages epidemiologists – who are mainly academics – to wittingly or negligently use various flawed statistical practices to produce positive, but (we show) likely false, claims.
There are numerous key biases that epidemiologists continue to unintentionally (or intentionally) ignore in studies of air quality and health effects. This is done to make positive, but likely false, research claims. Some examples are:
- multiple testing and multiple modeling
- omitting predictors and confounders
- not controlling for residual confounding
- neglecting interactions among variables
- not properly testing model assumptions
- neglecting exposure uncertainties
- making unjustified interventional causal interpretation of regression coefficients
Our study focused on the multiple testing and multiple modeling bias to assess whether a body of research has been affected by flawed statistical practices. We subjected research claiming that PM2.5 is harmful to a series of simple but severe statistical tests.
3. How epidemiologists skew research
Our study found strong circumstantial evidence that claims made about PM2.5 causing mortality, heart attacks and asthma are compromised by flawed statistical practices. These flawed practices make the research untrustworthy as it favors producing false claims that would not reproduce if done properly. This is discussed further below.
Estimating the number of statistical tests in a study – There is known flexibility available to epidemiology researchers to undertake a range of statistical tests and use different statistical models on observational data sets. The researchers then can select, use and report (cherry pick) a portion of the test and model results that favor a narrative.
One form of simple but severe testing we employed was counting. Specifically, we estimated the number of statistical hypothesis tests conducted in 70 different published epidemiology studies that make PM2.5−health effect claims. These results are presented in our study. The counting procedures are straightforward, and readers can learn and use them to count statistical tests in published observational studies. In our case, the median number of statistical tests performed in these 70 studies was over 13,000.
Epidemiologists typically use a Relative Risk (RR) or Odds Ratio (OR) lower confidence limit > 1 (or a p-value < 0.05) as decision criteria to justify a significant PM2.5−health effect claim in a statistical test. However, for any given number of statistical tests performed on the same set of data set, 5% are expected to yield a significant, but false result. A study with 13,000 statistical tests could have as many as 0.05 x 13,000 = 650 significant, but false results!
Given advanced statistical software, epidemiologists today can easily perform this many or more statistical tests on a set of data in an observational study. They can then cherry pick 10 or 20 of their most interesting findings and write up a nice, tight research paper around these findings—which are most likely to be false, irreproducible findings. We have yet to see an air quality−health effects study that reports as many as 650 results. How exactly is one supposed to tell the difference between a false positive or a possible true positive result when so many tests are performed and so few results are presented?
Diagnosing evidence of publication bias, p-hacking and/or HARKing – Publication bias is the failure to publish the results of a study unless they are positive results that show significant associations. P-hacking is reanalyzing data in many different ways to yield a target result. HARKing (Hypothesizing After Results are Known) is using the data to generate a hypothesis and pretend the hypothesis was stated first.
It is traditional in epidemiology to use confidence intervals instead of p-values from a hypothesis test to demonstrate statistical significance. As both confidence intervals and p-values are constructed from the same data, they are interchangeable, and one can be calculated from the other.
We first calculated p-values from confidence intervals for data from meta-analysis studies that make PM2.5−health effect claims. A meta-analysis is a systematic procedure for statistically combining data from multiple studies that address a common research question—for example, whether PM2.5 is a likely cause of a specific health effect (e.g., mortality). We looked at meta-analysis studies claiming that PM2.5 causes: i) mortality, ii) heart attacks and iii) asthma.
We then used a simple but novel statistical method—p-value plotting—as a severe test to diagnose evidence of publication bias, p-hacking and/or HARKing in this data. More specifically, after calculating p-values from confidence intervals we then plotted the distribution of rank ordered p-values (a p-value plot).
Conceptually, a p-value plot allows us to examine a specific premise that factor A causes outcome B using data combined from multiple observational studies in meta-analysis. What should a p-value plot of the data look like?
- a plot that forms an approximate 45-degree line provides evidence of randomness—supporting the null hypothesis of no significant association between factor A & outcome B (Figure 1)
- a plot that forms approximately a line with slope < 1, where most of the p-values are small (less than 0.05), provides evidence for a real effect—supporting a statistically significant association between factor A & outcome B (Figure 2)
- a plot that exhibits bilinearity—that divides into two lines—provides evidence of publication bias, p-hacking and/or HARKing (Figure 3)
Figure 1. P-value plot of a meta-analysis of observational data sets analyzing associations between elderly long-term exercise training (factor A) and mortality & morbidity (injury) (outcome B); data points drawn from 40 observational studies.
Figure 2. P-value plot of a meta-analysis of observational data sets analyzing associations between smoking (factor A) and squamous cell carcinoma of the lungs (outcome B); data points drawn from 102 observational studies.
Figure 3. P-value plot of a meta-analysis of observational data sets analyzing associations between PM2.5 (factor A) and all−cause mortality (outcome B); data points drawn from 29 observational studies.
We show over a dozen p-value plots in our study for meta-analysis data of associations between PM2.5 (and other air quality components) and mortality, heart attacks and asthma. All these plots exhibit bilinearity!
This provides compelling circumstantial evidence that the literature on PM2.5 (and other air quality components)—specifically for mortality, heart attack and asthma claims—has been affected by statistical practices that have rendered the underlying research untrustworthy.
Our findings are consistent with the general claim that false-positive results from publication bias, p-hacking and/or HARKing are common features of the medical science literature today, including the broad range of risk factor−chronic disease research.
4. How government bureaucrats skew policy
The process is further derailed with government involvement. The EPA have relied on statistical analyses to show significant PM2.5−health effect associations. EPA bureaucrats who fund this type of research depend on regulations to support their existence. The EPA has slowly imposed increasingly restrictive regulation over the past 40 years.
However, the EPA appears to have acted selectively in its approach to the health effects of PM2.5. This has been done by paying more attention to research that supports regulation (i.e., shows significant PM2.5−health effect associations) and ignoring or downplaying research that shows no significant PM2.5−health effect associations. This latter research exists, it is simply ignored or downplayed by the bureaucrats! Nor are the researchers finding negative results funded.
It is apparent to us that bureaucrats lack an understanding of, or willfully ignore, flawed statistical practices and other biases identified above in PM2.5−health effects research. They, along with environmental activists, continuously push for tighter air quality regulation based on flawed practices and false findings.
5. Can this mess be fixed?
Epidemiologists and government bureaucrats collectively skew results of medical science towards justifying regulation of PM2.5, while almost always keeping their data sets private. Far too many of these types, and a distressingly large amount of the public, believe that academic (university) science is superior to industry science. However, as epidemiology evidence is largely based on university research, we should treat it with the same skepticism as we would industry research.
Mainstream media appear clueless and uninterested in glaring biases in epidemiology research that cause false findings—flawed statistical practices, analysis manipulation, cherry picking results, selective reporting, broken peer review.
Epidemiologists, and government bureaucrats who depend on their work to justify PM2.5 regulation, proceed with far too much self-confidence. They have an insufficient sense of the need for awareness of just how much statistics must remain an exercise in measuring uncertainty rather than establishing certainty. This mess plagues government policy by providing a false level of certainty to a body of research that justifies PM2.5 regulation.
In our study we make several recommendations to the Biden administration for fixing this mess. However, we do not hold our breath that they will be considered. Some of these include:
- the administration needs to support statistically sound and reproducible science
- unsound statistical practices silently supported by the EPA need to stop
- the building and analysis of data sets should be separately funded
- these data sets should be made available for public scrutiny
Most importantly, Americans need to be aware that current statistical practices being used at the EPA for setting policy and regulations are flawed and obviously expensive.
S. Stanley Young (email@example.com) is the CEO of CGStat in Raleigh, North Carolina and is the Director of the National Association of Scholars’ Shifting Sands Project. Warren Kindzierski (firstname.lastname@example.org) is an Adjunct Professor in the School of Public Health at the University of Alberta in Edmonton, Alberta.
 Young SS, Kindzierski W, Randall D. 2021. Shifting Sands: Unsound Science and Unsafe Regulation. Keeping Count of Government Science: P-Value Plotting, P-Hacking, and PM2.5 Regulation. National Association of Scholars, New York, NY. https://www.nas.org/reports/shifting-sands
Pingback: How epidemiologists try to fool us with flawed statistical practices - HootNeoos
Thanks for an interesting post.
Thank you. I hope you get a fair hearing with the Biden Admin. If not, you will be, at the least, successful at putting some doubt in their minds where there may have been none. Consider House or Senate testimony.
Good start for our Freedom Fighters!
Willard … here’s a link to a webinar for shifting sands:
Here’s David Randall:
Being part of a conservative think tank doesn’t make you a bad person, any more than a liberal one. But … we do need to see the authors’ paper, which they said was available.
More generally, Bill, X does not imply Y whenever X does not entail Y.
For instance, the fact that Warren has been recruited by a fossil fuel company to pad its lobbying does not make Warren a bad guy:
That just means we’re dealing with a boringly familiar Freedom Fighter.
Willard, sometimes you can judge a book by its cover. Or, at the least, make some appropriate inferences. I get it. But many times, particularly in the present culture, that is taken too far. Remember that guy … “Are you now, or have you ever, been a member of the Communist party?” It doesn’t matter whether McCarthy was ‘right or wrong’. It was his methods, which were ironically UnAmerican, as freedom of association is a basic right. Your original observation that the link supposedly sharing the authors’ paper doesn’t exist is correct. They need to remedy it.
I just noticed the same and posted a duplicate comment.
First in the list of supporters for NAS, are
Anonymous Foundation A
Anonymous Foundation B
We can’t take poor wee willies words at face value. Contrast preserving life, liberty and the pursuit of happiness with his devotion to the AI economic overlord. Talk about tedious.
But the term is replicate – there is a so called crisis in replication that science is itself recognising and addressing. And apart from anything else the physiological effects of PM2.5 are known from animal models.
CliSciFi is not working on its problems. Just the opposite; it obfuscates, denies, and outright lies. Just read anything by its apologists.
There is lots of good climate relevant science.
PM2.5… reading the statistics on that form of air pollution may be overblown but hard to accept because, sounds too correct to dismiss.
Not sure that irreproducibility is necessarily the core issue here since it appears to assume that the same analysis on the same or similarly constructed data set would produce a different result. This may be the case – but that does not strike me as the main issue. I need to think harder about the authors methods for spotting issues. I have just been reading Koonin book “Unsettled” and one piece in the chapter on sea level rise struck me: Koonin noted: the senior authors of the CSSR report had never been asked and apparently had not considered the range of rates of sea level among earlier decades.. If there is no open review process of data and methods, errors are bound to made – better and sounder statistical methods are surely required but openness to contrary viewpoints strikes me as critical. .
Let us not forget Rud Istvan’s magisterial review of Steve Koonin’s new book.
Rud said it better than I did. The statistical errors addressed in this post are merely a symptom and not the real sources of the problems.
Bernie 1815 may have met his Waterloo:
Istvan’s review commenced by getting the one word title of Steve’s book wrong, and went downhill thereafter.
” If there is no open review process of data and methods, errors are bound to made – better and sounder statistical methods are surely required but openness to contrary viewpoints strikes me as critical.”
At the risk of piling on, this the point of the politicization of science. The lack of review is intentional, these are not “errors,” they enjoy the lack of sound statistical methods. They are telling a story, not doing science.
And it’s not just climate. Anthony Fauci demanded every vaccinated person wear two masks and remain outdoors 10 feet away from anyone else right up to the moment when some Republican SOB ridiculously asked “why?” Then the whole tower of cards collapsed in a single afternoon. One minute the CDC said science insisted you must wear two face diapers alone in your car, the next minute the CDC said science permitted you to attend packed movie theaters without a mask.
Our resident warmists are just the climate version of the folks with four doses of vaccine and three masks vigorously hand sanitizing alone in a field mumbling that everyone else is a dangerous denier. And if you don’t believe it, they have a chart of India with carefully selected dates to show you.
A compelling summary of your investigations. Thank you very much.
President Dwight D. Eisenhower warned the public about science capturing policy and vice versa. Where bureaucratic careers and Federal funding dollars are on the line, everybody is motivated to get “the right answer.” Ideological and political agendas also skew scientific investigations and public dissemination of preferred information (propaganda). This is also true in the highly politicized climate science arena (CliSciFi).
Because of pressure to toe the political line (lie) I quit Federal executive work after 11 years of progressive advancement. Having had an extensive career in electric power generation, transmission and distribution (planning, finance, design, construction and operation and maintenance), I have the knowledge and experience of a utility CEO/GM to tell you the Green New Deal is an economic and technological impossibility.
Please read up on some of the critical analyses of the politicized ‘science’ coming out of the CliSciFi/political complex. Just a few of the fine authors are Dr. Roger Pielke, Jr., Stephen Koonin and Marc Morano, and the websites “What’s Up With That” and “Climate Etc.” are excellent sources of technical information.
I’m guessing you would enjoy reading a website that specializes in articles about scientific study retractions:
After observing climate “science” for the past 24 years, COVID “science” in the past year, and reading RetractionWatch.com, I have come to the following scholarly conclusion about scientists, that could get me nominated for some kind of prize:
Never buy a used car from a scientist !
Stanley and Warren –
> Climate Etc. recently carried several insightful posts about How we fool ourselves.
I’d say thst one of the most common ways that people fool themselves is when they think that they get get into other people’s heads and mind-read and judge other people’s motivations or intentions.
> How epidemiologists try to fool us with flawed statistical practices.
Looking beyond the absurd generalization about “epidemiologists,” I’d suggest thst you have no valid way to assess epidemiologists’ intentions and/or motivations.
I think you should check yourself for filling yourself. You are the easiest person for you to fool.
Please note that the authors’ repeatedly said that the practices could be unconscious.
Josh’s contributions here are almost all quotes out of context or obscure “contradictions” that are perfectly consistent.
Most of these “biases” are no doubt unconscious even though the pattern is often so strong that conscious misrepresentation becomes a credible idea. The biases are usually based on cultural narratives or memes that most scientists don’t question, in some cases due to funding reasons.
How do you unconsciously try to fool people?
Some common meme’s
1. If I run the code right I will get the right answer since I have all the physics in the model.
2. If I get a bad result, it is due to bad gridding or other inputs or my lack of experience, because meme #1.
3. Researcher field leader X got good results so I should be able to get as good or better results. If I don’t my funders might be concerned.
4. Science gives us predictable theories and methods that if properly applied should give us a reliable answer.
> If I run the code right I will get the right answer since I have all the physics in the model.
> Josh’s contributions here are almost all quotes out of context
Lol. Quote out of context.
It’s the freakin’ title of the post.
“I’d say that one of the most common ways that people fool themselves is when they think that they get into other people’s heads and mind-read and judge other people’s motivations or intentions” writes the sometimes vociferous Joshua.
For sure life is always a dangerous game if and when we want to or try to get into other people’s heads, but it is something most of us stop trying to do once we have experienced mature and meaningful relationships/ friendships. We begin to appreciate that others do not always agree or see things the way we do and we accept this independence as a gift rather than a dangerous thing.
Alternative ideas or beliefs allow us freedom to explore and perhaps come up with original thinking of our own. We may then begin to understand and appreciate why free thinking people naturally resist groupthink, consensus, and censorship especially when they have unacceptable agendas. It is the mavericks who resist and expose deception as history tells us.
When scientists or researchers use specific means to demonstrate a theory then we should expect, at the very least, that they honestly understand the material they are using because if they don’t then they are treating everybody else as bigger fools than themselves.
I think actually it’s pretty important to try to get into other people’s heads in a sense, to understand how they reached the conclusions they reached, but to try to do so without interpreting their motivations or intentions.
I’m talking about something along the lines of “cognitive empathy,” or “perspective taking” – or more generally, simply not assuming that someone else is wrong, or malign in their intent (for example that they’re “try[ing] to fool us”), but trying to see why their view might be logical and stemming from intents similar to ours even, if they reach different conclusions than we do. Engaging with such a process can help us to see the impact of biases in reasoning, and in particular our own biases on our own reasoning.
‘Engaging with such a process can help us to see the impact of biases in reasoning, and in particular our own biases on our own reasoning.’ writes Joshua.
I look forward to the day that happens to you, Joshua, and will be delighted to let you know if I see it.
We all have a responsibility to value integrity and to be open, honest, and accepting of alternative theory regardless of peer review or whatever else society decrees as its methodology. It cannot help matters when there is a heavy bias towards one side of a controversy, especially where the bias is based upon an original and proven lie. With both Covid-19 and climate change we have seen the damage and harm ‘trying to get into people’s heads’ actually does. It isn’t healthy.
This is great work, thank you. It is very important because most recent emission reduction regulations are “beneficial” because of the PM2.5 health impacts using the epidemiolgist’s flawed numbers.
There is another aspect to PM2.5 health impact claims that I have yet to see. Ambient concentrations of PM2.5 are down everywhere in the country so it should be possible to validate the health impact claims by seeing an improvement in health outcomes. If anyone has seen such a study please let us all know. I looked at data for New York City here https://wp.me/p8hgeb-nC.
This doesn’t surprise me at all. This kind of thing is very common in even more rigorous fields of science. The one I know is virtually universal is selection of good results and ignoring bad results. This means the CFD literature for example is dramatically biased in a positive direction. The Navier-Stokes codes and methods are just flawed for all but the easiest cases and the whole field is not addressing it. Ditto for climate science and its addiction to models that are wrong about virtually all the patterns that matter.
The way to reform this is for funders to require registration of methods and outcomes before a study is funded. Also there needs to be a requirement to report all the data, not just your final result.
One of the reasons reform is so hard is that the political left denies there is a problem and has adopted “follow THE SCIENCE” as the new religious requirement. We have our local doofuses here too who don’t know enough to be credible but who try to deflect and discredit by ad hominem attacks. Anonymity is their white robe and hood.
David Young says –
dpy6629 | April 4, 2021 at 12:46 pm |
You are cherrypicking Josh as is often the case. If you look at the Wiki page on the Swedish epidemic cases and admissions are indeed both rising but deaths were declining throughout March.
7-day moving average of daily deaths in Sweden.
March 1, twenty deaths per day.
April 19, twenty deaths per day
Josh is obsessed with proving DPY WRONG
May 15th – 7 day moving average is 14 deaths per day. Looks like a declining death rate to any rational observer (non-obsessed observer).
Sweden’s current death rate (7day moving averating is in the range of 0.2 per 100k and has been at that range since early february 2021. That death rate puts Sweden among the lowest death rates in the world.
Truly interesting that you still haven’t incorporated the time lag in accounting for deaths into your thinking about COVID outcomes in Sweden (IOW, you need to go back considerable farther than the 15th to make a confident assessment), but yes, given the data on case numbers and ICU admissions, it is reasonable to expect that the death rate will now be going down – as opposed to the time when David Young said it would be going down – at the end of February – despite me telling him that he was wrong and that he needed to take the lag into account.
In response to me telling him that, he called me a liar and insulted me in any other manner of ways. What’s intersting is that his error came despite his being a renown scientist l, and me being a non-scientists. The point being, his appeals to (his own) authority were only evidence of the fallaciousness of appeals to authority.
And meanwhile, another beautiful irony is that you’re comparing death rates in Sweden to the death rates in other countries, without even a passing reference let along a serious consideration of the impact of confounding variables when assessing the efficacy of different mitigation or interventional policies.
That’s ironic because of the theme and focus of this post.
At any rate, in contrast to the broadscale characterizations about epidemiologists thrown around by the authors of this post, in my observation the field of epidemiology actually generally takes the issue of differentiating between associations and causality pretty seriously as compared to some other fields of science. The adherence to the principles of Hill’s Criteria is one common practice I see in epidemiology that reinforces my view, as does the standard practice of epidiologists to include a “limitations” section in their publications – which often discuss cautions and caveats not to assume cauality from cross-sectional or observational data, or not to generalize from non-representative sampling.
At any rate, Joe, given your consistent failure to discuss the importance of accounting for confounding variables when discussing the comparative pluses and minuses of Sweden’s and other countries’ COVID policies, you might benefit from re-reading this post.
Yes Joe, Josh can’t say anything important so he focuses on minutae. You will notice that the word count for his comments is huge given the lack of content. You will also notice that he gives no time when he assembles his cherry picked “numbers.” April 4 would the relevant date since that’s when he quotes me. It’s a hopeless mass of errors and emotion.
I’ve learned that one cannot have reasoned discussions with the Joshua’s of the world. Don’t feed their delusions.
David Young –
On April 4th, and other days around that time, I explained to you that your statements that the daily death rate in Sweden were significantly declining were wrong because it was clear that you hadn’t accounted for a long-established pattern of a lag in the death reporting.
Here’s just one example of many such exchanges:
> dpy6629 | April 5, 2021 at 10:36 pm |
Josh, I’m talking actual numbers. you are talking unquantified generalities, so typical of your unfocused intellect. ICU admissions have been rising almost exactly tracking cases. It’s actually been 7 weeks since Feb. 1. That long a lag is very unlikely. If I grant you 2 weeks, which once again you have no evidence for, that still 3 weeks overdue for the increase in deaths given the widely acknowledged 2 week expected lag.
You were wrong to claim that deaths were leveling off and you cover that with more very repetitious word salads, another sign of an unfocused intellect.
It’s interesting that a data scientist would make such an obvious error and that a non-scientist could see your error so easily. If course that’s not a reflection of something particularly significant about me, as your error was indeed so obvious.
It’s also interesting that in response to me pointing out your error, you thought it was appropriate to insult me so many times and in so many ways.
Joe is right Joshie, its mostly irrelevant and the fact that you continue to return to it ad infinitum tells everyone here your time is not valuable to you and that my characterization of your “contributions” is correct.
Josh – you are spending too much time trying to account for confounding factors and the minutia without grasping the broader picture.
Sweden moving 7day average death rate normalized by population has been on a very steady decline since February 2021 to the point that Sweden’s average death rate is among the lowest in the western world.
Your constant repeating – “7-day moving average of daily deaths in Sweden.
March 1, twenty deaths per day.
April 19, twenty deaths per day”
While your correction is true – in the very broad analysis it is trivial and meaningless.
It’s a simple point.
Back in early April David Young said that while cases were increasing in Sweden, rate of ICU admissions and rate of deaths were “strongly declining.”
I pointed out that he was obviously wrong on both counts. I explained that ICU admissions had in fact strongly increased, and that while deaths had decreased significantly up until the end of February, from that point forward (if you account for the lag) the rate of deaths had flattened out.
After which David Young insisted he was right I was wrong, and piled on with a whole series of insults. What makes it even funnier is that David Young then went on to appeal to his own authority as some kind of evidence that he was right, and my not being a scientist as evidence for why I was wrong.
No, isn’t a significant issue – and it’s far from the only time that David Young has made similar errors. In fact he made a whole string of such errors.
But it’s an interesting issue (to me at least) in that David Young repeatedly demonstrates a whole series of the very same bad arguments and fallacious reasoning discussed in this post and indeed, the whole series of posts here about how and why people make bad arguments.
Now you are certainly entitled to not find thst I interesting, but then it’s curious why you tho k it’s worth commenting about. I wondeid it’s related to your tendency to comment about COVID outcomes in Sweden by making comparisons to other countries without assessing the impact of confounding cseia led in the conclusions you’ve drawn?
The broader point here is that Josh is behaving like a typical propagandist. He can’t respond to the point of this post so he tries to discredit by various meaningless references. Out of tens of thousands of my comments, he found one in which he can lie about the data as it existed on April 4 and use much later data. At the time it was said it was true. The curve was monotone decreasing.
Joshie has a history of obsession with his intellectual betters. I am among good company as it includes Nic Lewis and Judith. Josh is simply an ankle biter who doesn’t realize how silly he looks. And then he finds truth telling about him to be insulting.
“Joshie has a history of obsession with his intellectual betters. I am among good company”
And humble with it.
What truly lacks humility is you and Josh, who lack real technical knowledge and rely on untrustworthy media such as Twitter, the New York Times, and the Guardian, but think they can fool people by coming here and misrepresenting what vastly more competent people say to discredit them. Your last foray on model accuracy was particularly disgraceful. You simply made up something that I never said and attributed it to me. You then later admitted I was right.
Because implementing a separated flow capability in TRANAIR gives you the authority to pontificate on every single scientific topic ever.
David Young –
> Out of tens of thousands of my comments, he found one in which he can lie about the data as it existed on April 4 and use much later data. At the time it was said it was true. The curve was monotone decreasing.
I didn’t “lie” about the data as it existed on April 4th. At that time I told you that you were wrong about your assertions about the death rate in Sweden, primarily because you were failing to account for the lag. I told you that at the time and you said I was wrong at the time.
You were wrong. I was right.
You made a whole series of erroneous statements related to the pandemic. Your statement that the death rate in Sweden was declining throughout March was just one. There was your incorrect statement about ICU admissions dropping when they were increasing.
There were your erroneous statements about the rates of testing and the rates of cases in Florida NY, and Texas respectively.
There were your incorrect statements about the pandemic ending last summer.
There were your incorrect statements about the spike in cases last summer being only a function of more testing or more younger people testing. You were wrong in your arguments that the spike in cases last summer wouldn’t lead to a parallel spike in morbidity and mortality.
And perhaps most amusingly there were your silly claims that Nic was wrong when he said that deaths in Sweden would top off at around 7,000, or that it was likely that nowhere would population fatality exceed 0.085%, 9r that Sweden would reach hers immunity close to a year ago, and again they’d reach it a few months ago, or that India, Geneva, and NYC had reached herd immunity months ago, etc.
Josh, You just lied in this thread by citing a death number for April 19 which is irrelevant to my assertion. The rest of your word salad is just the result of an unfocused and undisciplined mind biting at people’s ankles.
No Willard, but it means I know how to read the literature unlike you who is playing a game here trying to cast doubt with irrelevancies. I don’t think anyone takes what you say seriously since you are an academic philosopher with no scientific qualifications and an obvious inability to read the literature and interpret it in a balanced way. What you have quoted is taken out of context and cherry picked.
BTW, I don’t remember saying that Nic was right about the ultimate death toll in Sweden and given your unfocused mind, I doubt if you know either. That’s called a misrepresentation Joshie and its your MO.
To prove this, I went back and found the figures in my comment of April 9 that prove you are lying about the data Josh. When you cite April 9 data its irrelevant and a logical error.
dpy6629 | April 9, 2021 at 2:31 pm |
Sweden’s population fatality rate is about 0.14% and is lower than many other European countries. Deaths continue to decline despite Joshua and your gaslighting this issue. The current curve on Worldometer is monotone decreasing. To say that it has plateaued is a falsehood. Monotone decreasing means that they continue to decline and were at 15 per day on March 31. They were at 19 on March 15 and at 21 on March 1 and 24 on February 15.
BTW, The data on April 4 was even lower.
“rely on untrustworthy media such as Twitter, the New York Times, and the Guardian,”
which, aside from being false we can note that our intellectual betters, with humility, like dpy, instead rely on reliable sources, such as…
Humility is as humlity does.
At least with CFD, you can generally go to a test facility and see if the model checks out with reality. Alternatively, can verify that the real machine behaves as predicted, although that approach could end up being expensive if the machine behaves badly.
The issue here is that there are billions of flow patterns we are interested in and each test costs a lot of money.
Pingback: How epidemiologists try to fool us with flawed statistical practices
While I have long been a reader of Climate Etc., WUWT, Climate Audit and The Air Vent in the early days, Jo Nova, Not a Lot of People Know That, Dr Roy Spencer and more, I have found it quite difficult to find good quality blogs expressing the Establishment view. RealClimate was a dud, banning me after the first try at commenting.
Any suggestions? I am still trying to work hot what motivated these people in the direction of poor science.
Also, as ever, it is easier to accept statistical methods that incorporate properly-calculated mathematical uncertainties of observations. Lack of these in the past has covered a multitude of sins. Geoff S
From teh wiki
Not to be confused with National Academy of Sciences (NAS).
The National Association of Scholars (NAS) is an American non-profit politically conservative advocacy group
Lewandowsky all the way down.
While the 3 National Academies are American nonprofit politically left wing advocacy group. On climate they are very alarmist.
Ground Level Ozone has been blamed as leading cause of premature mortality with several studies supporting that claim. The study “Ozone and Short-term Mortality in 95 US Urban Communities, 1987-2000” is considered the gold standard for studies of this type.
Yet this study considered to be the “gold Standard” suffers from so many basic errors that the conclusions should be considered meaningless
1) lack of controls
2) biased data gathering
3) levels of ground level ozone in several cities so low that the ground level ozone cant possibly be a factor.
4) much higher correlations of other factors.
5) negative correlations of ozone with premature deaths in relation to correlations with other factors.
I will add that the EPA’s study/Studies used for the 2.5pm analysis appears to suffer from similar shortcomings.
Milloy has a whole book on the PM2.5 hoax:
One example of the short sightedness of the EPA came to light with their campaign to reduce ground level ozone during the early 1990’s. The EPA required the reduction of products emitting VOC’s (Volital organic compounds). Oil based paints were required to reduce their VOC’s from 320 to 280 per gal. The result was that the adhesion and durability of the paint dropped considerably. The effect was a trivial reduction in ground level ozone in the short term, more frequent painting which actually creates the release of greater amounts voc’s in the long term, and a corresponding increase in solid waste (pollution). (note that I am making no comment regarding the health hazard of ground level ozone – just noting that the EPA decision results in greater pollution, not less)
My observation of the EPA’s and the studies associated with the PM2.5 is that the EPA concluded that the law of diminishing returns was repealed – possibly by the obama administration
That’s interesting. I Googled CGStat and got this.
Politicized science can’t be fixed, it can just be exposed. Embarrassment and humiliation are the tools to delegitimize government agencies like the EPA, CDC, etc. Then take away their power, that’s the only fix. Politicized science provides useful cover to politicians and regulators to legitimize regulations, as in providing observational studies and intuition to support mask requirements to prevent the spread of Covid-19, while serious science (even when published in a CDC journal) is ignored, because it is not useful. A collection of 75 years worth of randomized controlled trials was analyzed, and systematic review found no significant effect of face masks on transmission of laboratory-confirmed influenza. https://wwwnc.cdc.gov/eid/article/26/5/19-0994_article
It took the CDC about a year to accept the obvious…that airborne transmission of COVID19 is a significant source of community transmission. Once you accept that then masks can’t make much difference. Unsurprisingly they haven’t made much difference.
I point at
(C1) airborne transmission of COVID19 is a significant source of community transmission
(C2) masks can’t make much difference
That is all.
Educate yourself on what airborne transmission means and what the masks we are wearing can and cannot do and it will all become clear.
Masks make a difference. See the link I posted on the pathological science regarding “aerosol” transmission. Basically, in this context [assuming the article is correct], they had the idea that only particles <= 1 micron floated in the air as opposed to being pulled down by gravity. A lot of people then fixated on the smaller possible size – a viron, which is way under a micron – and concluded masks won't work. Of course, a single viron is also unlikely to cause an infection.
But that idea is wrong. There is a whole spectrum of particle sizes, from visible droplets to maybe even individual virons. A wide range of sizes, far larger than 1 micron, will float considerable differences. Masks will block these to a significant amount. A lot of masks work for much smaller particles – particles significantly smaller than the mask mesh, even.
But there has been a mess in messaging, and masks became political.
See my comment below in response to you.
> Educate yourself on what airborne transmission
You were saying?
You were saying….
“…measures to avoid inhalation of infectious aerosols, including ventilation, air filtration, reducing crowding and time spent indoors, use of masks whenever indoors….”
This is why earlier this month:
schools made children wear masks while outdoors and running alone.
Vice President Harris, while vaccinated, made a point of wearing a mask on television, outside, while kissing her husband, who was also vaccinated.
And Willard posts a paragraph noting that all of that was complete bunk, while claiming it says the opposite.
Meanwhile, Josh posts “daily death” averages in Sweden that are lower than the daily average of deaths in New York City- which is and was locked down and double-masked during the same period. This also is intended to prove, “scientifically,” that the lockdowns and mask mandates in NYC “worked” unlike the absurd freedom in Sweden.
And then, in the space of 24 hours, the CDC reversed all the guidance because… they ran out of excuses to go along with the partisan nuts.
> This is why earlier
I’ll take that squirrel as a win, JeffN.
Well its not a very clean story on masks. In a community setting there is no real evidence of effectiveness for the flu. In hospitals N95’s do help but compliance tends to be low because they are uncomfortable and have side effects.
But the available evidence seems pretty sparse and not really conclusive, like so much in viral epidemiology. It’s really a field where it is mostly meaningless to refer to “THE SCIENCE.” That politicians do it means they are ignorant of the real science and its lack of conclusiveness.
Particularly outdoors, there is no reason for anyone to wear a mask unless they are in a densely packed group for a long period of time, like at a BLM mostly peaceful riot.
> Well its not a very clean story on masks
Well I suppose that when one is OK with selling military weapons anything goes.
Who is selling military weapons?
Pingback: How epidemiologists try to fool us with flawed statistical practices – Climate- Science.press
Has this been peer-reviewed? Submitted for publication? Is the data/software all available?
This link gets me page not found.
I recommend the following recent, related article because it discusses what seems to be a major flaw in the infectious disease world regarding airborne transmission. And that flaw, if the article is accurate, is a result of pathological science processes – mostly the sociology, not fraud or bad intentions – that I believe cost a whole lot of lives in this epidemic.
TL;DR: the distinction between aerosol and droplet spread is muddled, and muddled in such a way as to disfavor accepting or understanding spread by airborne (not falling) particles/droplets/whatever.
Although not mentioned IIRC in the article, this same muddle has confused the argument about masks, since it includes the incorrect idea that aerosol spread (non-falling particles) is only by particles <= 1 micron. That fools people, especially non-specialists, into believing that masks won't help because it would seem that the particles will get through them.
As for the main post, the organization (NA Scholars) certainly has a bias, and "try to fool us" in the title is inappropriately pejorative. But most organizations, including major scientific ones, have biases, even if they are not formed for that purpose. It's natural. The various pathological science issues they raise are real, and quite common in many fields, and continue to be even with more and more articles and research papers pointing out that fact. I have no idea how accurate their criticism is, but attacking it just based on their political stance is employing the ad hominem fallacy.
Particles can get through the masks we all wear, but that is not the only issue. The masks we wear do not seal and have gaps that allow significant bypass flow…and that is if worn properly. Frequently we do not wear them properly and we do not wear them at all in the places airborne transmission is most likely to occur.
Once we accept airborne transmission as a primary path of transmission then what is important changes drastically. Masks and social distancing mean far less than air turnover in enclosed spaces, UVC sterilization, and/or filtration with filters that can remove small particles, e.g. HEPA filters.
A century of real world data demonstrates that mask wearing by the general population doesn’t work. This pandemic has just reaffirmed that earlier work.
Lots of theories.
Funny how the same contrarian concerns can be peddled whatever the issue.
Let’s think about designing an experiment to test the efficacy of masks. We can’t treat a mask like a HEPA filter, which is used on mechanical devices for filtration. There we could position a sensor for the particulates to pass around or through, or use a membrane which could catch the particles for later analysis. It would seem we need to use a manikin-type head for which the mask is physically designed. Maybe we would hollow out a section behind the mask and place our sensors/membrane?
But that wouldn’t quite replicate a desirable portion of reality. Certainly nowhere near what we would expect for a mechanical device, such as a fan. Why? I wore N95s, half-face with different cartridges and other sundry masks for almost 3 decades in industrial and commercial construction. My particulates of concern were solids not liquids. However, my observations can be applied to both. Aside from the general differences in facial configuration between races, the differences within races is astounding. We all use our facial muscles differently, we all perspire differently. And all masks when worn over time require adjustments.
I know that many of you reuse the same mask for a long time. I used to see how the dirt would collect around the material pressed against my face. How long does it take for the mask to reach a saturation point where the particulates move past it and seek entry along the contact area?
My point, sorry for the delay, is that I believe there is no way to prove a mask works with any reasonable quantification. Yet, if you ask me if they ‘worked’ for me, I’d say they did a pretty good job … for my type of particulates. So, it may come down to personal choice. The government should layout the choices, the potential benefits to wearing a mask along with the fact that there is no scientific proof. If you think they are a must, then by all means wear one. Or two, if that makes you feel comfortable. Making it mandatory gets into a different ball game … a political game.
There can be debate about why masks don’t work to prevent community transmission of disease, but the evidence is overwhelming that they don’t. People act as if there have been no studies that investigated this issue. There were studies published as far back as 1919 and 1921 that investigated mask effectiveness following the 1918 pandemic and many studies since then. You would be hard pressed to find a study before March of 2020 that did not reach that conclusion. Most studies during this pandemic have been correlation studies that cherry picked data. The one RCT performed for this pandemic demonstrated no protection to the wearer. We were supposed to believe that somehow they still protect those that aren’t wearing them. And because transmission is airborne there is no reason to expect them to be effective at preventing transmission. They are not positive fit test masks they are pieces of cloth with significant bypass flow around them.
> There can be debate about why masks don’t work to prevent community transmission of disease..
There are many problems with your thinking on mask usage, but this is probably the best way to see your biggest problem.
No one is looking for masks to “prevent” disease from spreading. The point of mask usage is maginal risk reduction of individual events of potential transmission, which then could compound across the population level.
Contrary to your description of the relevant literature, research findings have been mixed about this issue – an issue which for obvious reasons is hard to investigate in real world context.
So it is an issue of decision-making about the possibly of mitigating perhaps low probability but high damage function risk – not all that unlike climate change as an issue.
> The one RCT performed for this pandemic demonstrated no protection to the wearer
Assuming you’re referring to the Danish mask study, that study didn’t investigate mask-wearing for source control, which is by far the main theorized benefit from mask wearing. It didn’t control for compliance (it was based on usage recommendations) it was underpoweeed and only looked for an effect of 50% or greater. Those are the reasons why the authors cautioned against using their study to push the kinds of conclusions you’re using the study to push.
There certainly is room to debate why troglodytes punch hippies.
Well, I don’t think an unmeasurable “marginal” reduction compounded across a population is worth a moment’s thought. Just like there might be an unmeasurable “marginal” reduction in cancer risk from high dose Vitamin C, no competant doctor can recommend it. It would be unethical to do so. Vitamin C curing cancer was another crackpot theory of Linus Pauling along with his passion for nuclear disarmament.
David Young –
> Just like there might be an unmeasurable “marginal” reduction in cancer risk from high dose Vitamin C,…
It seems that perhaps you don’t understand what individual marginal risk benefit compounding at the population level means. That wouldn’t happen with vitamin C usage. It has the potential to happen with mask-wearing. Do I really need to explain this to you?
Compliance is a legitimate area of debate IMO, but so what do we plan to do fund mask police? You act like I referenced that study alone when I referenced it in the context of a century of other studies.
>… so what do we plan to do fund mask police?
Doing a highly effective real world mask study would be extemely complicated, not just expensive.
The question I think is more important is how do you make decisions given that you have highly imperfect information.
In a situation of an essentially constant rate of infection, the population wide effect is just the sum of the individual effects just as with Vitamin C. In a case of exponential growth there might be a bigger effect but these waves don’t last more than a few months anyway so its a totally speculative idea.
In any case, Josh, you cited no scientific evidence there is any effect at all.
> It seems that perhaps you don’t understand what individual marginal risk benefit compounding at the population level means. That wouldn’t happen with vitamin C usage. It has the potential to happen with mask-wearing.
Maybe you need to flesh that out a bit more, for me at least. I see that concept as being far from certain. What comes to mind is the 55 MPH speed limit. It sounded like a policy of individual marginal risk benefit compounding at the population level. Yet, when speed limits were raised the number of deaths didn’t go up, even though they initially went down. Obviously other variables were at play.
Mask wearing shouldn’t be confused as a slam dunk, nor is risk analysis. It’s a potential guide based on partial/incomplete information. There’s utility in mask use, but mask use alone would not defeat the virus. The problem is it can’t be quantified.
David Young –
> In a situation of an essentially constant rate of infection, the population wide effect is just the sum of the individual effects just as with Vitamin C.
Hard to believe you actually need an explanation.
If one infected person wearing a mask reduces the number of people he/she infects in a supermarket by just one (say zero people infected out of hundreds in the market rather than one person infected), that one person differentially not infected could infect zero members of his/her pod that he/she spends lits of times indoors with rather, than perhaps two or five or ten. Then that fewer number infected, might infect fewer other people in their own overlapping pods. And so on.
It’s a hypothetical effect Joshie since there is no evidence for an effect in the first place. You have cited nothing and as usual wasted people’s time with hypotheticals that in no way contribute to constructive discussion.
> I see that concept as being far from certain.
I didn’t say it’s “certain,” or a “slam dunk.” My point is that you have to make these assessments in the face of uncertainty. I’ve said that multiple times.
The marginal benefit of lower individual risk from speed limits doesn’t compound at the population level as it does with mask wearing – except in the sense that one person less likely to get in an accident due to a lower speed limit is less likely to injure one more more people in that accident that was prevented. But the risk benefit doesn’t compound across the population level as with mask wearing.
David Young –
> It’s a hypothetical effect Joshie since there is no evidence for an effect in the first place.
Good. I think you understand now., and see why your vitamin C comparison is basically a non-sequitur.
Since your a renown scientist I’m surprised you needed an explanation from an inferior such as myself.
Yes, the effect is theoretical in the sense that the exact individual maginal risk benefit isn’t proven. But there is solid evidence to suggest a marginal benefit exists and this is about the evaluation of policies to address risk in the face of uncertainty.
No Joshie, You have shown nothing to show that a “marginal” benefit exists. You are just speculating based on nothing as usual and wasting poeple’s time.
> I didn’t say it’s “certain,” or a “slam dunk.” My point is that you have to make these assessments in the face of uncertainty. I’ve said that multiple times.
Noted. Those were my words. And I agree that the risk assessments need to be made. My issue is with taking assessments and presenting them as policy mandates. I would prefer they be presented as choices, respecting a person’s right to decide for themselves, even though these choices affect the society at large. I think we agree on the general utility of masks, but it’s our reliance of social mechanisms for getting the appropriate use that differs.
Thanks for your responses.
Information to a free people vs bureaucratic dictates.
> I would prefer they be presented as choices, respecting a person’s right to decide for themselves, even though these choices affect the society at large.
And others, often those who are most vulnerable, would prefer them as mandates, under the thinking that your focus on “freedoms” likely puts others at a loss of “freedom” to move relatively less exposed to unnecessary risk. In many countries the public doesn’t have the view that there’s a zero sum relationship between individual rights and “choices” and the common good. They think that an expectation that people will wear masks is a basic societal responsibility. They think that Americans are self-centered and foolishly focused on confusing a trivial inconvenience with “freedom.”
Joshua, you obviously don’t understand the Constitution of the United States of America, especially its first Ten Amendments (Bill of Rights). In the USA, the individual’s natural rights are supreme, not the government dictates. Citizens are encouraged to voice their objections to governmental intrusions outside of Constitutional limitations on governmental power.
Individual States exercise what may be called ‘police’ powers that provide for laws to further regulate lawful behavior. Those, however, are also limited by the Constitution. Ultimately, “social responsibility” is too nebulous a concept to be dictated by central authority. Let society primarily take care of itself outside of the legislative and judicial arenas, although there are obvious exceptions. We have always been able to muddle through on our own.
> And others, often those who are most vulnerable, would prefer them as mandates, under the thinking that your focus on “freedoms” likely puts others at a loss of “freedom” to move relatively less exposed to unnecessary risk. In many countries the public doesn’t have the view that there’s a zero sum relationship between individual rights and “choices” and the common good.
And because there are differences of views as to social utility, whether individual or common good, it isn’t a zero sum game. It’s not black and white, a yes or no, all or nothing issue. That’s what makes democracy ‘messy’.
And the term most vulnerable is a subjective term, particularly when it comes to policy. And that’s leaving aside the subjectiveness of remedies. I understand, and applaud, your passion for applying science to solve human ills, I just want to be careful that once the ill has passed that we have a society that hasn’t been damaged. There should be attention to both. And that level/balance is attained through our political values.
The American citizens who want to exercise their natural right to limit their unnecessary risk of morbidity and mortality want you to stop. whining about a minor inconvenience, stop fantasing that you’re Paul Revere because you whine about a minor inconvenience, and wear a mask under limited circumstances.
Joshua, I had WuFlu in July, 2020 and was fully vaccinated earlier this year. Government dictates cannot consider such individual conditions. If you are fearful of the China virus, do what you want. Otherwise, leave me alone.
I agree with your 11:41 pm commemt.
We found that cotton masks, surgical masks, and N95 masks all have a protective effect with respect to the transmission of infective droplets/aerosols of SARS-CoV-2.
Just to continue a bit with Americans, social constructs like ‘most vulnerable’, perceptions of our fellow citizens’ behavior, … you might find some things in this article interesting.
This caught my attention: “The thought of conflating your entire political, moral, social, family, and religious being with your professional persona,” Garcia-Martinez says, “I think is extraordinarily fraught and difficult.”
It not only results in tyrannical behavior towards others, it’s much more sadly a tyranny to ourselves, as this layer upon layer of ‘politically acceptable terms/positions’ act like a social burka to separate, stigmatize and categorize, ultimately preventing us from experiencing each other’s humanity. We lose the ability to relate to each other, as we have subsumed individuality to categorizations.
Willard’s reference is largely meaningless here. It deals with simulations not actual real world trials. The reasoning is that masks can block some particles therefore they must “work.” It’s fallacious reasoning. Cancer chemo drugs kill lung cancer cells in the lab, therefore chemo should work for lung cancer. Except double blind trials show the effect is really small. In fact, there is far too much of this kind of fallacious reasoning being promoted by activists. Crude mechanistic narratives are prefered by activists like Willard, but they are not real actionable science.
“The reasoning is that masks can block some particles therefore they must “work.” It’s fallacious reasoning. Cancer chemo drugs kill lung cancer cells in the lab, therefore chemo should work for lung cancer. Except double blind trials show the effect is really small.”
No, your reasoning by analogy is fallacious. Cancer is a disease with extremely rapid mutation, while SARS-CoV-2 is not.
If masks block some of the particles, then they reduce transmission when they are worn. That is obvious, and is backed up by research with other diseases, and the hamster experiments with this disease. One is not going to get double blind realistic trials without actual challenge experiments, and the ethics of doing that for masks are very questionable. Otherwise, you have to go with lab experiments on non-humans, and experience with other transmissible diseases where masks are shown to work – such as SARS.
> It deals with simulations not actual real world trials.
I too would welcome real world trials.
Srsly, David, you’re just saying stuff.
David Young –
I’m surprised to see such a renowned scientists employ such fallacious reasoning.
First you fail to understand why, if masks do reduce transmission, the effect of individual risk reduction would compound at the population level (your misunderstanding was made obvious by your irrelevant analogy to vitamin C).
Then you fail to understand why masks don’t have to be 100% effective at preventing all particles from passing through in order to reduce infections (your misunderstanding was made obvious by your reference to chemotherapy).
What’s next, you’re going to say that COVID deaths in Sweden “strongly declined” throughout March because you don’t even account for such an obvious factor as the lag in reporting?
dpy6629 | April 4, 2021 at 12:46 pm |
You are cherrypicking Josh as is often the case. If you look at the Wiki page on the Swedish epidemic cases and admissions are indeed both rising but deaths were declining throughout March.
Josh, You generate word salads better than anyone else here. The main fact here is that the best science shows masks in a community setting don’t do anything to reduce transmission. You have contributed nothing to anyone’s knowledge or understanding here.
The analogy is still apt even though not perfect. The main point is that what “works” in the lab often doesn’t work in the real world. In that regard chemo and masks are alike.
I am going to stop responding to you as it really is a waste of everyone’s time and clutters up Judith’s threads with the meanderings of a confused and unfocused mind.
Well Meso, the papers I’ve seen don’t show a significant effect of masks in a community setting at least with the flu. There is an effect with hospital workers of N95 masks perhaps because they might tend to wear them properly and they are tight fitting (and also have side effects). Generally loose fitting surgical masks just deflect the air upward or to the side. In an indoor setting, I don’t think that’s of much benefit because the ambient air will rapidly become uniformly contaminated because of convection. In an outdoor setting, conditions are more hostile to viruses especially when the sun is out. I’ve seen little evidence for significant outdoor transmission but if you have some, I’d be interested.
“Masks in community settings have no clearly proved efficacy (9–14). In three trials, participants were randomized either to hand washing or to hand washing plus surgical masks (9, 11, 13), with no clear additional benefit of masks. ”
I’ve also seen some recent papers analyzing European data vs. timing of mask mandates. As I recall there was little correlation but I don’t have time to dig it up right now.
Speaking of cherryicking, David, the paper you cite has been received July 03, 2012.
If you click on the “cited by” tab, here’s the first hit you get:
You’re a joke.
David Young –
> Well Meso, the papers I’ve seen don’t show a significant effect of masks in a community setting at least with the flu.
The evidence related to mask usage and COVID is mixed. Even a brief examination of the literature will make that obvious. It’s strange that you looked at the literature yet don’t realize that.
Also, as I have already explained to you, even a marginal benefit in reducing the risk of transmission from individual interactions would compound at the population level. Masks don’t have to reduce transmission 100%, or even 50%, to have a significant public health impact.
Why do you have such difficulty understanding that? It’s curious that a renowned scientist would struggle with such a basic concept.
Well Willard, I’m afraid your crude and totally superficial textual analysis is the only joke here.
I don’t trust anything on masks or interventions coming out of the science “establishment” the last year and a half. These are the people who said BLM riots were fine but churches were not. Thousands of people packed into a small space, shouting, and without masks mostly. Covid has been by far the worst episode of total politization of science at least since the 19th century Progressive craze for Social Darwinism and racism. The CDC has flip-flopped on it several times with Fauci at one point saying you don’t need to wear a mask.
Masks is a very complex subject. There are many many different types and most people don’t wear them properly anyway. N95’s do make a quite significant difference in both medical and industrial settings, but they are uncomfortable and have medical side effects such as chronic skin rashes, shortness of breath, or even hypoxia. It’s hard to get even medical personnel to fully comply with their use.
Let’s bite and find the next meta-analysis in the same “cited by” list:
Willard, you are a cherry picking machine, aren’t you? Your citation also says at the end of the abstract.
“This systematic review and meta-analysis supports the use of respiratory protection. However, the existing evidence is sparse and findings are inconsistent within and across studies. Multicentre RCTs with standardized protocols conducted outside epidemic periods would help to clarify the circumstances under which the use of masks or respirators is most warranted.”
The abstract also seems to conflate different types of masks and resperators giving a single number for all types, which strikes me as not good science.
So be it:
Masks work. Don’t be silly.
David Young –
> These are the people who said BLM riots were fine but churches were not. Thousands of people packed into a small space, shouting, and without masks mostly.
Apparently despite being a renowned scientist (who constantly appeals to his own authority), you’re not aware that churches mostly meet indoors.
Remarkable that you wouldn’t even nconsider that.
Although with so many other failures in your understanding of basic logic, perhaps not surprising
Pingback: How epidemiologists try to fool us with flawed statistical practices – Watts Up With That? | Blue Anon News
Pingback: How epidemiologists try to fool us with flawed statistical practices – Watts Up With That?
They’re so successful with Operation COVIDIUS that today they feel empowered to do even more…
People tend to spend 85% to 95% of their lives indoors. So any study of pollution that ignores indoor pollution is not complete. And that probably means nearly every study !
Here are some quotes from an April 2019 article I wrote on the subject:
Paula Olsiewski, a biochemist, asked Delphine Farmer, a chemist based at Colorado State University, to develop new instruments and data on indoor atmospheric chemistry.
Farmer recruited Marina Vance, and they gathered 20 research groups, from 13 universities, and launched “Homechem”, a large field study.
Homechem used the “UTest House”, a twelve-hundred-square-foot prefab house that cost sixty thousand dollars.
At first, the instruments, originally designed for outdoor atmospheric
measurement, had to be re-calibrated to deal with much higher concentrations of chemicals that build up indoors.
According to the World Health Organization, India’s capital city, New Delhi, has THE WORST outdoor air quality of any major city.
During the dirtier Indian winter months, fine particulate matter levels are about 225 micrograms per cubic meter.
That’s not much different than the 280 micrograms per cubic meter that was reached during the final hour of cooking a big Thanksgiving dinner inside a test house! ”
It’s a good point. Indoor air pollution in the third world caused by indoor fires is the worst problem. One easily solved by introducing fossil fuels.
I spent six weeks in Northern India Oct-Nov 2017. There was a constant layer of pollution perhaps 40-50 feet up, with no wind or rain to disperse it. I needed anti-biotics for much of the time there. Much of the layer, I think, was from the wood, dung etc fires used for cooking and heating.
Just happened to come across again this paper showing that there is a almost a huge positive results bias in the literature and one way to address it.
But scientologists and other outsiders with political motivations will continue to say we should “follow THE SCIENCE.”
/ Do you realize
That everyone you know someday will die?
And instead of saying all of your goodbyes
Let them know you realize that life goes fast
It’s hard to make the good things last
You realize the sun doesn’t go down
It’s just an illusion caused by the world spinning round /
The link to the report on NAS web site doesn’t work and the listing of reports from the menu doesn’t show this report.
Did the report get pulled? Lack of scientific rigor perhaps? Bad data?
Very true… remember Michael Mann lost his multi million dollar lawsuit against Tim Ball because he would not release his hockey stick data or R2 regression numbers. More Data needs to be public
Concerning the Mask & Distancing controversies:
“Filtration Efficiency …”
JAMA Intern Med 2020;180(12):1607-1612
.”…with procedural face masks secured with elastic ear loops showing the lowest efficiency (38% overall fitted filtration efficiency).” [the paper ones most people wear]
“Two metres or one: what is the evidence …?”
BMJ 2020;370:m3223 doi: 10.1136/bmj.m3223
The study Willard posted above showed similar results. Efficiencies of what we are wearing of between approximately 40-60% in a test chamber where the “subjects” are about 19″ apart. This, and others, taken with the real world data that masks don’t make a noticeable difference in community transmission is a completely consistent result.
> don’t make a noticeable difference
Keep doubling down, Doug:
At some point you are going to have to learn to deal with the top down result. Masks do not impact community transmission.
It travels more than 2m based on talking and breathing also. You obviously still do not understand what airborne transmission means.
I think I’m going to read you every paragraph that mentions masks until you cry uncle, Doug:
If masks were not effective, low compliance with mask wearing would make no difference. That’s obviously not what the authors presume.
You’re trying to deny stuff people know since the black plague.
“This, and others, taken with the real world data that masks don’t make a noticeable difference in community transmission is a completely consistent result.”
Huh? A 40%-60% reduction in transmission means a very substantial reduction in the R value.
The problem with your community transmission comment is measurement. We don’t have good experiments, because they measure mandates, not actual behavior. What we’ve seen is that mask wear will go down over time, because people get tired of it. But mask wear can and probably did make a significant difference in transmission.
In a situation where no vaccines will exist, in some sense, all you’re doing is postponing the inevitable, although mitigations do appear to have prevented hospital overload in developed countries, which saves a lot of lives. But with vaccines coming, or now, being available, mitigation measures make tons of sense until they aren’t needed. And, at least in the US, the case levels are dropping low enough, while the sum of vaccinated and previously ill unvaccinated is likely at or above the herd immunity threshold. We will see – as things relax more – whether the threshold is higher and the drop is due to a combination of immunity and mitigations.
As I have stated repeatedly, at some point people are going to have to accept the top down result. Literally dozens of studies demonstrate that masks don’t matter in real world transmission of ILI. Why that is is irrelevant. My opinion is that there is no reason to expect them to work considering the imperfect fit and significant bypass flow. Who nows, maybe if we all just wore them properly all the time they would help but they don’t help based based on real world data across multiple studies and across more than a century.
This is a seasonal respiratory infection behaving as a seasonal,respiratory infection behaves. That is at least part of the reason it is waning in the NH. Logically and hopefully some is also due to vaccinations but we will have to wait and see how much. The 7 day rolling average death toll in the USA was about the same as it is now in July of 2020 months before anyone had been vaccinated.
The statistics issue in climate science
Pingback: Irreproducible science and US government regulation – Watts Up With That?
Pingback: Irreproducible science and US government regulation – Climate- Science.press
As I’ve mentioned before, community settings are very hard to measure, and thus the studies I’ve seen have been far from convincing.
That’s why people study the physics of mask wear and transmission mechanisms. Those studies show clear benefit from mask wear. Likewise, hospital studies show that, including a study showing that surgical masks were as effective as N95 for SARS in a health care setting.
Masks are shown to reduce transmission, which should be obvious anyway – they reduce the odds, from any situation, of a transmission. That is equivalent, population wide, of reducing the Rt value.
Over a long term, without improvements in care or vaccines, all that would do is potentially reduce medical system overload, because herd immunity would have to be reached by natural infection. But that isn’t the situation with COVID19 – care has improved and we have remarkably effective vaccines.
And, very importantly, masks cannot be the only mitigation unless the disease has a low R0, and they may not be the appropriate mitigation in other circumstances – mitigation involves multiple choices and combinations.
Well I actually mostly agree with this comment Meso. My issue is just whether the effect is measurable and significant in many settings. Outdoors I doubt if masks make a difference unless you are at a BLM riot and packed in a crowd shouting at the top of your lungs for hours. On trails and in parking lots, I just don’t think you are likely to get a significant number of viral particles. The viral concentration goes down like the square of the distance at least not accounting for particles falling to the ground.
In medical and industrial settings N95 masks really do help a lot if worn properly. I doubt if surgical masks make much difference even indoors. They mostly deflect the air flow up or to the sides. This air will diffuse throughout the room pretty quickly so the air will have about the same level of viral particles as if you were not wearing a mask.
“Well I actually mostly agree with this comment Meso. My issue is just whether the effect is measurable and significant in many settings. Outdoors I doubt if masks make a difference unless you are at a BLM riot and packed in a crowd shouting at the top of your lungs for hours. ”
We are in agreement, mostly, on this. I almost never wear a mask outdoors. I would note that, while we now know how important aerosols are, droplets still count, if you are close enough.
The problem with “measurable” is that we don’t have the ability to measure it, for reasons previously mentioned. That’s why looking at the physics is so important – i.e. lab experiments. Lots of science is done in labs, including biological science, even though the results are applicable outside the lab.
One problem I have seen, especially in the west and with WHO, is public health folks – in this case, epidemiologists and other specialist – demanding solid scientific results before acting. That sort of institutional bias, and habit, served us very badly during this epidemic.
If you can’t measure it well, that doesn’t mean it doesn’t exist. It just means you can’t measure it. And that’s a major reason for lab work. After all, it’s pretty hard to catch a lot of subatomic interactions in the wild, but physicists don’t say that since they only see them in exotic accelerators, they don’t matter.
SARS required close contact transmission. Indicating droplet transmission was a more important mode than it is for COVID19. Masks have always been shown to be more effective in that case. If close contact transmission was required for COVID19 it would never have become a pandemic IMO. Just as SARS and MERS never became pandemics.
“I don’t think that means what you think it means.” Who, what, when and where are the real questions.
More on the replication crisis.
“ Nonreplicable publications are cited more than replicable ones”
We use publicly available data to show that published papers in top psychology, economics, and general interest journals that fail to replicate are cited more than those that replicate. This difference in citation does not change after the publication of the failure to replicate. Only 12% of postreplication citations of nonreplicable findings acknowledge the replication failure. Existing evidence also shows that experts predict well which papers will be replicated. Given this prediction, why are nonreplicable papers accepted for publication in the first place? A possible answer is that the review team faces a trade-off. When the results are more “interesting,” they apply lower standards regarding their reproducibility.”
Has that study been replicated?
It makes sense. More sensational studies will be more likely to be cited and will also be less likely to be correct.
If the efficacy of Ivermectin in treating COVID-19, suggested by many peer-reviewed scientific papers, is borne out, YouTube will be revealed to have been playing a very dangerous game indeed. How much health—individual and economic—will be sacrificed globally on this altar? It’s censorship in science’s clothing. Look closely, and you will find that this has little to do with science. A censor wearing a lab coat is still a censor, and censorship is fundamentally incompatible with science.
This sounds familiar. It sounds like a slam of consensus science to me. When she brings together censor and science I think she’s reducing it to its starkness. Look in the mirror. Is that what we’re supposed to do?
This stupid game when lives are at stake, in the near and long term.
What do you think, in retrospect, about the many times you rejected my comments pointing out that your comments about the likely trend in COVID infections and deaths in Sweden would be wrong?
Also, are you aware that Ivermectin was being taken rather commonly in India, that there was a marked increase in infections and death regardless, and that large #’s of people taking it off-label and w/o clinical supervision has significantly complicated the treatment of such patients?
Josh comment – “Also, are you aware that Ivermectin was being taken rather commonly in India, that there was a marked increase in infections and death regardless …”
Josh – that marked increase in infections and deaths in India remain significantly below virtually every industrial country.
“What do you think, in retrospect, about the many times you rejected my comments pointing out that your comments about the likely trend in COVID infections and deaths in Sweden would be wrong?”
Ragnaar is probably ignoring you comments regarding Sweden since those comments are pointing out the minutia without looking at the broader picture. For example the overall trend in Sweden has been declining since January, There have been several short term upticks during those 5 months but the overall trend has been a declining trend, albeit at times slow decline, but a decline never the less.
Yes Joe, Joshie always focuses on minutae because his mind is an unfocused and undisciplined mess. He’s purely derivative as well in that his comments always focus on finding irrelevant ‘errors’ in other’s more interesting contributions.
> Josh – that marked increase in infections and deaths in India remain significantly below virtually every industrial country.
The question is why has the infection rate increases so dramatically recently despite widespread use of Ivermectin.
The per capita rate is one thing and the rate of relative change is another. Have. You looked at the rate of change there?
> Ragnaar is probably ignoring you comments regarding Sweden…
Ragnaar made many, many comments about the future trajectory of the pandemic in Sweden, again in a relative sense, as a reflection of his views about the advisability of their covid policies.
Quite frequently, he posted charts of the rate of infections in Sweden to suggest that their policies had kept the relative rate of infections tamped down, that they wouldn’t have a second wave.
But when their rates exploded and they not only had a large 2nd wave but also a large 3rd wave he never came back to reflect on his failed analysis.
I was wondering what he had to say about that.
Here’s one example. I’ll look for more if you still don’t get that Ragnaar was speaking to the issue I was speaking to
Ragnaar | December 11, 2020 at 9:20 pm |
We have potentially good news:
“This means that for the last 10 days of data, death counts in Sweden must only be interpreted as incomplete measures of mortality.”
I’ve been watching the 10 day lag deal. We are now three or four days past 10 days. The full definition says more than 10 days. But a reasonable person would think the further back in time we go, the less the chances of getting adds.
We have been hoping for a leveling off. It may not be here. What we don’t seem to have is exponential growth which would be bad. It’s had plenty of time to ignite. That a country hanging by a thread would have gone up in flames by now. It hasn’t.
Everything will Okay.
Joshua | December 11, 2020 at 9:58 pm |
Their ICU admissions rate has starte to level off. But it sure ain’t dropped. And it’s still going up.
But…they’ve run out of room in their ICUs. They’re passing laws so they can shut down businesses. They’ve just shut down their middle schools. Why did they shut down their middle schools?
You still don’t understand the lag. All this time and you still don’t get it. Must be a reason.
They’re still adding deaths to over 17 days ago. Why are they still adding deaths to over 17 days ago? Yesterday they added deaths to their 7-day average for deaths on the 26th of November they added deaths today. It’s still going up. They are now listing 135 deaths between the 24th and 25th. Check back next Tuesday to see if it goes up again.
Why do you never learn?
Why do you not care that other people sacrifice? That other people’s family members die? Why don’t you care? I like to think it’s motivated reasoning. Maybe it’s Trump cultism. Maybe it’s just some twisted form of hatred of libz and demz and “the left. ” Because even that’s better than the alternative explanation.
Josh – you are spending too much time obsessing over the minutia. Sweden death rate was approx 1 per 100k during nov/Dec / Jan. In Feb/March/April / May , the death rate is down to .1 to .2 per 100k. Who cares if you found a Gotcha for a bounce up to .2 from a .1. Who cares that Sweden has a reporting lag that cause .05 per 100k deviation from actual. ‘Who cares about all the confounding factors, those confounding factors are trivial when viewed from a higher level.
Again, the overall infection rate and death rate have dropped – dropped considerably. The short term bounces are trivial.
Josh’s comment (in relation to India) – “The question is why has the infection rate increases so dramatically recently despite widespread use of Ivermectin.
The per capita rate is one thing and the rate of relative change is another. Have. You looked at the rate of change there?”
Go to 91-dovic for a good graph of the trajectory of covid in each country.
Regarding India –
1) you should note that India for the most part is just now going through their first wave. For all practical purposes, India escaped the first wave during 2020.
2) Being a country located south of the 30thN and north of the 30thS, according to the Hopes simpson curve, India should be experiencing a wave during this time of year. That wave is not surprising. the sharp uptick would be expected. You should note that western Europe and the US experienced a large second wave during oct/nov/Dec/Jan . That large uptick should have been expected.
3) India’s uptick over the last couple of months is running 20-25% of western europe & the US infection and death rates (comparatively with the respective waves). Why is India’s uptick so much less than the US and western Europe – possibly because ivermectin.
Taking ivermectin wouldn’t seem to have any impact on infection rates, but potentially on death rates.
“Rob Starkey | May 23, 2021 at 1:53 pm |
Taking ivermectin wouldn’t seem to have any impact on infection rates, but potentially on death rates.”
Sorry for any confusion – The purpose of comment was to convey a relative sense of covid spread in India vs the US and Western Europe.
Many comments have pointed to India’s “exploding covid infection and death rate. People making those comments apparently are unaware 1) the India’s death and infection rates and running 20%-25% of most other western countries and 2) unaware that India if for the most part just now going through their first wave.
Just trying to convey a broader perspective
Are you comfortable comparing the infection and death rate of different countries given their different data collection processes?
I was wrong about Sweden and I applaud them for their success. Leave it to Sweden to show us how to hang on to individual liberties. Holding true while being the subject of criticism put us to shame with our Fauci worship.
With 221 deaths per million, India also put us to shame. They must be better educated than us. They outlasted the British didn’t they? Less french fries per capita too.
“What if there are long-term effects of mRNA vaccines?”
Low cost hedges. I have some stocks and some bonds. You take the vaccine, I’ll take Ivermectin. They work towards the same goal in different ways. Sweden does it one way, we do it another. Africa has to beg for help and New Zealand goes more nuts than they were before.
Our government spent the past decade losing its credibility and continues that trend now. The same government that tells me men are women needs to be listened to when I still have a choice?
I see you’re posting my previous comments. Thank you. Sweden’s daily death count can be described as having a barbell distribution. The first bell was supposed to limit the second bell. I expected a diminishing returns plot which is common where? In nature. So what is the cause of the second bell? Whatever the answer is, it involves many variables and getting to more than three of them throws us into increasing uncertainty.
The answer can be posed as either nature or man. If the virus exploited Sweden to return with a second bell, at least we learned something. And experimenting with a small country that has resources is not the worst thing that could’ve happened.
Data can be used to blame or to learn. Addressing your remarks from December, I applaud Sweden for being an adult and not the clown show of the United States. They retained things that we seem poised to lose. And they’re leftists.
I applaud Sweden for not cowering and taking it out on their children. For retaining more of their freedoms and individual liberties. They showed the way for our Red States who showed our Blue States how to do it.
The data says, our Blue States failed, same as always. Some people are nuts and can’t see it.
> I was wrong about Sweden and I applaud them for their success…
Looking past the way in which you define success (obviously it would depend on the criteria chosen, which will vary by person).
That you were wrong isn’t particularly interesting to me. Errors are to be expected under conditions of high uncertainty. I was interested to see if you might touch on the question of WHY you were wrong, repeatedly, and why even as I explained (in advance) to you why why you were likely going to be wrong, you continued with the same erroneous expectation about future trends in Sweden.
I’m also hoping that if you listen to this, you’ll rethink your apparent belief that weinstein and, Heying are credible:
> Brett Weinstein & Heather Heying: Why are ‘they’ suppressing Ivermectin, the miracle cure?
Also this other podcast – are you aware that they’ve claimes that three of the four Weinstein children have been unjustly denied Nobel Prizes?:
> Brett Weinstein & Heather Heying: Why are ‘they’ suppressing Ivermectin, the miracle cure?
You should listen to their Darkhorse podcast. They addressed that very thing. It’s a lame article and if you read it, you would not have brought it up. I support Peterson don’t forget, and I’ve seen the grasping attacks against him. The first article I linked on this subject sums up Heying’s position well. They haven’t claimed it’s a miracle cure. Jesus. They are taking it which is more than follow through than any of their critics are likely to be able to demonstrate.
People like you used to be wary of big pharma. For good reason. They capture the regulators like any decent crony capitalist. This isn’t rocket science.
You should also have a look at the Darkhorse show notes where they back things up as scientists sometimes do. And while your at it, you can ponder a Patreon donation for their efforts.
Thank you J.
Upon further review of your Darkhorse Podcast critics we find they describe themselves as:
A Northern Irish cognitive anthropologist who occasionally moonlights as a social psychologist. Chris has long standing interests in the psychology of conspiracy theorists and pseudoscience. His academic research focuses on the Cognitive Science of Religion and ritual psychology.
An Australian psychologist and numbers-guy. He does research on all kinds of stuff, but particularly enjoys looking into why people believe the things they do: religion, conspiracy theories, alternative medicine and stuff.
That they’ve locked on to Weinstein and Heying show just how much Kool-Aid they’ve drunk. Conspiracy theories?
The data at the link below show the total (all-causes, all ages) deaths in Sweden per 100,000 population by month from 2015-2021.
The black bar is 2020 and the impact of covid-19 is clear.
The red bar is 2021.
This is an excellent and very detailed analysis. Age and population adjusted mortality in Sweden for 2020 was up only a little bit and was comparable to 2013. It’s too detailed for Josh to follow but others will find it interesting.
Since this is being lied about elsewhere the original source of this information is the software development perestroika website which is non political and uses official government statistics. Their posts are very complete and discuss uncertainties a lot.
Trying to link this to a site that merely reprinted it is a sham and a misrepresentation. Typical of unethical concensus enforcers.
“Rob Starkey | May 23, 2021 at 2:20 pm |
Are you comfortable comparing the infection and death rate of different countries given their different data collection processes?”
For discerning general trends and relative differences in infection and death rates at a broader/high level – yes. For purpose of discerning the minutia, then no.
India may data collection may be understating the infection rate by 30-50%, If so, then India’s infection and death rate is running 30%-40% below the US and western Europe,instead of the 20%-25% less than reflected in the 91-divoc data / john hopkins data.
I’d agree with your overall analysis of the trend in India. I’d be far less comfortable comparing India’s data to another country and drawing conclusions about best practices.
““Rob Starkey | May 23, 2021 at 2:20 pm |
Are you comfortable comparing the infection and death rate of different countries given their different data collection processes?”
For discerning general trends and relative differences in infection and death rates at a broader/high level – yes. For purpose of discerning the minutia, then no.
India may data collection may be understating the infection rate by 30-50%, ”
Some country infection rates are just not very useful. Consider Mexico – take a look at 91-divoc. They look pretty good. But they’ve had a terrible crisis there, ongoing. Many people cross the border into here in Arizona just to get a COVID19 vaccine, and many more to end up in our hospitals getting care for it.
Here’s the blog with the best analysis of Swedish data. It shows that things were not nearly as bad as everyone said.
Can’t that data be highly misleading.
Isn’t it more important to review how a particular locations health care system was able or did respond to the pandemic over time.
Was the Swedish healthcare system degrated due to being over capacity? Were shutdowns to help with healthsystem capacity concerns or other motivations.
I don’t know the answer. I have heard that Sweden has lower ICU capacity than most other advanced countries. My brother tells me though that it’s easy to turn a regular hospital bed into an ICU bed. The staffing is the challenge. Hospital workers get sick too or stay home when they are scared.
Pingback: Weekly Climate And Energy News Roundup #456 – Watts Up With That?
Pingback: Weekly Climate And Energy News Roundup #456 – Climate- Science.press
One of the ways we fool ourselves is the idea that Global warming caused by Global emissions has an implication for non global climate action heroism of nation states.
Pingback: El Cambio Climático existe y es producto de un capitalismo reaccionario, el Pacto Verde es la no-solución del capital • Communia
Pingback: Climate Change exists and is a product of reactionary capitalism, the Green Deal is capital's non-solution • Communia