Barry McGee avatar

Barry McGee

Software Engineer

Thoughts

RT @GabrielScally: The island of Ireland is in trouble. The population isn't safe unless #COVID19 is controlled across all of the island. N…

In the week we're being asked to think about celebrating partition, can anyone look at these stats and tell me how partition isn't responsible for people needlessly dying during this pandemic? https://t.co/hBXK2XCKsK

RT @laasrinadiaa: The definition of happiness for developers: - Closing 50 chrome tabs after solving a bug 😅 What's your definition?

@tj_ohanlon Firstly, that is not an article I asked for about global deaths vs p-value. Secondly, that article is about the significance threshold when you can’t prove a negative in an experiment. That plainly doesn’t apply to death as you can prove the negative - the person still alive.

@tj_ohanlon Scientists debate that global deaths should be used instead of a P-Score? Can you provide a link please?

@tj_ohanlon Picking one year as an example and picking one year in isolation is the same thing. You’ve just shared another total deaths chart when I’ve already said it’s the P-Score that’s statistically significant. C19 deaths against most countries population totals will be a rounding error

@tj_ohanlon Picking one year in isolation is statistically misleading. You need to look at P-Score - excess mortality as the % diff between the number of weekly deaths in 2020 and the average number of deaths in the same week over the previous 5 years. It's bad. https://t.co/cxMe24s9uj

@tj_ohanlon But the previous table you shared shows an upward tick on the fifty year trend, even before covid?

@tj_ohanlon Every single year since 1950 that number has incremented down until 2019/20 when it has went up? But.. that's positive?