It's flagged because it's a random dude from Twitter's dubious interpretation of a graph.
Interpreting raw data isn't something you can do to draw conclusions if you're not a domain expert. From the replies below the tweet, what's being claimed and what the data says isn't certain. Moreover, I'm not hearing from domain experts being concerned about this in the least.
So, no, it's just that showing a random graph and drawing your favourite conclusions just isn't something worth posting.
This question clarifies what exactly is the denominator. It is the population of that age group and is not stratified further by vaccination status. So this will naturally bias as more people get vaccinated within an age group.
This is an incorrect interpretation: "If I am an individual who is not vaccinated I will be less likely to be infected than someone my age who is vaccinated."
This (I think) is a fair interpretation: "If I am an individual who is infectious I am more likely to be vaccinated." While technically correct, how meaningful is it on it's own when we know more people are in the vaccinated group?
To roughly compare outcomes within an age group between vaccinated and unvaccinated people you'd need to additionally account for proportion of vaccinated status at that date. I may followup with those adjustments attempted.
I think its possible that the OP twitter author may be pointing out that there are issues of diminishing returns and questions of whether a mandate should be supported when you are more likely to be exposed to the virus by someone vaccinated vs not. I don't think I can quite tell if this argument stands on its own but I think there are far greater reasons to not support mandates.
4 comments
[ 3.0 ms ] story [ 15.6 ms ] threadInterpreting raw data isn't something you can do to draw conclusions if you're not a domain expert. From the replies below the tweet, what's being claimed and what the data says isn't certain. Moreover, I'm not hearing from domain experts being concerned about this in the least.
So, no, it's just that showing a random graph and drawing your favourite conclusions just isn't something worth posting.
This question clarifies what exactly is the denominator. It is the population of that age group and is not stratified further by vaccination status. So this will naturally bias as more people get vaccinated within an age group.
This is an incorrect interpretation: "If I am an individual who is not vaccinated I will be less likely to be infected than someone my age who is vaccinated."
This (I think) is a fair interpretation: "If I am an individual who is infectious I am more likely to be vaccinated." While technically correct, how meaningful is it on it's own when we know more people are in the vaccinated group?
To roughly compare outcomes within an age group between vaccinated and unvaccinated people you'd need to additionally account for proportion of vaccinated status at that date. I may followup with those adjustments attempted.
I think its possible that the OP twitter author may be pointing out that there are issues of diminishing returns and questions of whether a mandate should be supported when you are more likely to be exposed to the virus by someone vaccinated vs not. I don't think I can quite tell if this argument stands on its own but I think there are far greater reasons to not support mandates.
Total UK pop in 30-39: 7.56 million (75.6 hundred-thousand)
(chart says vaxxed ~775/hundred-thousand unvaxxed ~750/hundred-thousand during final week)
Final week infections in age-group 30-39 had 58600 in the vaxxed and 56700 in the unvaxxed.
Now renormalize to the subpop (vaxxed vs unvaxxed) This age group was 76.5% vaccinated in the last week).
Vaxxed infections/100k = 58600.76575.6) = 1011 Unvaxxed infections/100k = 56700(1-.765)75.6 = 3203
So if you are in this age group during the latest released week you'd be 3x more likely to be infected if you were unvaccinated.