Ok #ygk, I have a serious thread coming to end this series.

Part 3 of 3 focuses on what happens to people after they turn away from the church.

Most said they were shunned, excommunicated, and defamed by their former friends, even by some of their family members when they left.

In part 1, I described how people’s whole lives played out at the church.
https://t.co/TiqrkjJONI

After leaving, they had to relearn how to live.

They never listened “secular” music or watched certain movies, some were even told to boycott Starbucks, because it supports Pride
Going back into the "real world" made one former member feel like she was "losing her mind."

Also, part 3 shows the extreme fear that some people had after leaving.

One person said she worried she would literally be struck by lighting if she spoke badly about the church. #ygk

More from Society

global health policy in 2020 has centered around NPI's (non-pharmaceutical interventions) like distancing, masks, school closures

these have been sold as a way to stop infection as though this were science.

this was never true and that fact was known and knowable.

let's look.


above is the plot of social restriction and NPI vs total death per million. there is 0 R2. this means that the variables play no role in explaining one another.

we can see this same relationship between NPI and all cause deaths.

this is devastating to the case for NPI.


clearly, correlation is not proof of causality, but a total lack of correlation IS proof that there was no material causality.

barring massive and implausible coincidence, it's essentially impossible to cause something and not correlate to it, especially 51 times.

this would seem to pose some very serious questions for those claiming that lockdowns work, those basing policy upon them, and those claiming this is the side of science.

there is no science here nor any data. this is the febrile imaginings of discredited modelers.

this has been clear and obvious from all over the world since the beginning and had been proven so clearly by may that it's hard to imagine anyone who is actually conversant with the data still believing in these responses.

everyone got the same R

You May Also Like