Attempts to ask these questions of various datasets are vitally important and you have found a really great resource to apply your questions, and to replicate other's findings. It is a lot of good effort. For which many are grateful, as there is clearly a danger signal for one or more shots.
Here are my impressions:
With all of these analyses published on a variety of Substacks, when trying to look at number of jabs, mortality and correlations, there is always the problem of sampling from the population without replacement. In other words, once you have killed off a number of people with the first shot, it is a different sample for the second shot. Some of the most vulnerable were taken out. Some of the people who took one shot and decided against the second might have had a bad reaction and decided, "whoa no more for me." Or maybe they started to see bad news trickle in. For example, my sister fortunately decided against the shot after seeing three people in her social and work circle promptly pass away after their one or two shots. At some point they stopped refrigerating the shots to be so cold. Other groups have verified bad lots and serious problems and variation in manufacturing. The same thing then happens again between shot 2 and 3 and 3 and 4 to further winnow and change the nature of the remaining sample. This would attenuate the correlation a bit for shot two, and then even more for shot three as you have now taken out more people from the same population who potentially had highest vulnerability to these shots if in fact that is a thing.
Ideally you want three separate populations to test, randomized or well-matched to compare separately 1, 2 and 3 shot schemes. Given the sampling without replacement and the other variables adding possible variation to the outcome including inaccuracies in reporting of shots it is remarkable that these signals are still so strong, as death is the dependent variable. The findings so far still supports the recommendation that all shots should be stopped immediately and all data sifted through.
Of course, all this work should have been done by pharma prior to FDA authorization, a randomized set of trials.
So there may be sophisticated techniques to help overcome, post hoc, these biases in the sampling. I think Matthew Crawford might have referred to progressive bias in sampling as survivor bias. There are also regression techniques in the area of survival analysis that might apply--who survives the shot schemes the longest, what makes them different using independent variables related to your overall theory and/or supplied by critics?
A comprehensive analysis with these sorts of multiple data sets across populations and the work of many thoughtful analysts is due--maybe as a coordinated roundtable with steering committee to plan and then conduct analyses. Also, if someone has access to graduate students in statistics or population public health data analysis it seems like this type of work would be ideal to that setting, as the project would be a layered, iterative, programmatic and complex set of questions. Such could be the focus of a master's or dissertation thesis with oversight of a large and qualified committee, including outside members who have been active on these questions and datasets. (I've been a member of about 2 dozen dissertation oversight committees and 4 master's theses).
Who wouldn't want "The Naked Emperor" as one of a committee to sign off on their thesis?
I'd recommend you make this article immediately free to all (currently has 7 day paywall which is fine on other articles for sure) and ask people to circulate on Social Media asking for evidence and analysis for support or refutation.
Mathew Crawford on Rounding the Earth has done similar work and has shown, at least to my satisfaction, that a lot of people who likely would not have died from covid died from the jab.
Really? I discovered him after he was attenuating Berenson's claims, and he came across like the jab was overall neutral. Albeit his stats expertise seemed convincing, I think he's prone to undercooking the damage of the "experimental injectables" he referred to them as.
EthicalSkeptic certainly has the expertise and isn't undercooking the damage.
I asked him directly on one of his recent deep dives on scatter graphs, would I be accurate in saying a lot of people who likely would not have died from covid died from the jab? He basically said yes.
Sep 7, 2022·edited Sep 7, 2022Liked by NE - Naked Emperor Newsletter
You should switch the axis. Normally you want the independent variable on the x-axis and the dependent variable on the y-axis. That’s pretty standard convention in the sciences. It’ll also help with bringing the intercept back towards 1 which is what you’d like to see if there is causality; which would bolster you’re case. Secondly, seeing the correlation decrease with an increasing number of shots could be evidence against the shots being the primary cause of excessive mortality. However if the slope increased with increasing shots, then that would be the evidence you are looking for. However, you need to switch the axis to calculate the right slope.
Looks much better. Also on reflection I’m not sure the slope needs to increase with additional shots. It would be better if it did, but for structural reasons in the data it is hard for to in practice. I’m curious what happens with a multi linear regression against all three percentages. If all three constants are positive and significant then that suggest additional risks for each additional shot.
The best fit changes when you switch the axes. The correlation stays the same. The slope doesn’t just invert it is also corrected by a factor of r^2. And the change to intercept is more complicated yet.
"Should not be ignored" but probably will be. The never changing always screaming retort, "correlation does not equal causation" I can already hear. The data, the deaths, the horrific adverse events numbers; nothing seems to stop the juggernaut of bad medicine.
Attempts to ask these questions of various datasets are vitally important and you have found a really great resource to apply your questions, and to replicate other's findings. It is a lot of good effort. For which many are grateful, as there is clearly a danger signal for one or more shots.
Here are my impressions:
With all of these analyses published on a variety of Substacks, when trying to look at number of jabs, mortality and correlations, there is always the problem of sampling from the population without replacement. In other words, once you have killed off a number of people with the first shot, it is a different sample for the second shot. Some of the most vulnerable were taken out. Some of the people who took one shot and decided against the second might have had a bad reaction and decided, "whoa no more for me." Or maybe they started to see bad news trickle in. For example, my sister fortunately decided against the shot after seeing three people in her social and work circle promptly pass away after their one or two shots. At some point they stopped refrigerating the shots to be so cold. Other groups have verified bad lots and serious problems and variation in manufacturing. The same thing then happens again between shot 2 and 3 and 3 and 4 to further winnow and change the nature of the remaining sample. This would attenuate the correlation a bit for shot two, and then even more for shot three as you have now taken out more people from the same population who potentially had highest vulnerability to these shots if in fact that is a thing.
Ideally you want three separate populations to test, randomized or well-matched to compare separately 1, 2 and 3 shot schemes. Given the sampling without replacement and the other variables adding possible variation to the outcome including inaccuracies in reporting of shots it is remarkable that these signals are still so strong, as death is the dependent variable. The findings so far still supports the recommendation that all shots should be stopped immediately and all data sifted through.
Of course, all this work should have been done by pharma prior to FDA authorization, a randomized set of trials.
So there may be sophisticated techniques to help overcome, post hoc, these biases in the sampling. I think Matthew Crawford might have referred to progressive bias in sampling as survivor bias. There are also regression techniques in the area of survival analysis that might apply--who survives the shot schemes the longest, what makes them different using independent variables related to your overall theory and/or supplied by critics?
A comprehensive analysis with these sorts of multiple data sets across populations and the work of many thoughtful analysts is due--maybe as a coordinated roundtable with steering committee to plan and then conduct analyses. Also, if someone has access to graduate students in statistics or population public health data analysis it seems like this type of work would be ideal to that setting, as the project would be a layered, iterative, programmatic and complex set of questions. Such could be the focus of a master's or dissertation thesis with oversight of a large and qualified committee, including outside members who have been active on these questions and datasets. (I've been a member of about 2 dozen dissertation oversight committees and 4 master's theses).
Who wouldn't want "The Naked Emperor" as one of a committee to sign off on their thesis?
I'd recommend you make this article immediately free to all (currently has 7 day paywall which is fine on other articles for sure) and ask people to circulate on Social Media asking for evidence and analysis for support or refutation.
I'd also get rid of the question mark in the title. You've shown the correlation, state it.
Or question the causation.
Agree.
Mathew Crawford on Rounding the Earth has done similar work and has shown, at least to my satisfaction, that a lot of people who likely would not have died from covid died from the jab.
Really? I discovered him after he was attenuating Berenson's claims, and he came across like the jab was overall neutral. Albeit his stats expertise seemed convincing, I think he's prone to undercooking the damage of the "experimental injectables" he referred to them as.
EthicalSkeptic certainly has the expertise and isn't undercooking the damage.
I asked him directly on one of his recent deep dives on scatter graphs, would I be accurate in saying a lot of people who likely would not have died from covid died from the jab? He basically said yes.
"This is how real science works"
NAKED EMPORER IS THE SCIENCE!
no, that is Fauci!
like
You should switch the axis. Normally you want the independent variable on the x-axis and the dependent variable on the y-axis. That’s pretty standard convention in the sciences. It’ll also help with bringing the intercept back towards 1 which is what you’d like to see if there is causality; which would bolster you’re case. Secondly, seeing the correlation decrease with an increasing number of shots could be evidence against the shots being the primary cause of excessive mortality. However if the slope increased with increasing shots, then that would be the evidence you are looking for. However, you need to switch the axis to calculate the right slope.
Thanks, I'll add the switched axis.
Looks much better. Also on reflection I’m not sure the slope needs to increase with additional shots. It would be better if it did, but for structural reasons in the data it is hard for to in practice. I’m curious what happens with a multi linear regression against all three percentages. If all three constants are positive and significant then that suggest additional risks for each additional shot.
Maybe standard but surely just translates the line!?
Related note.
I would say keeping same scale across the charts helps
The best fit changes when you switch the axes. The correlation stays the same. The slope doesn’t just invert it is also corrected by a factor of r^2. And the change to intercept is more complicated yet.
"Should not be ignored" but probably will be. The never changing always screaming retort, "correlation does not equal causation" I can already hear. The data, the deaths, the horrific adverse events numbers; nothing seems to stop the juggernaut of bad medicine.