https://www.graphsaboutreligion.com/p/political-ideology-religious-attendance
Just a few observations myself.
The first table separating atheists and agnostics is insightful. Generally surveys throw atheists and agnostics together and for good reason since they are very similar in social attitudes. But there are very real difference here, particularly among conservative atheists and agnostics.
it is already well established in the literature that progressives have lower levels of well-being than conservatives. But that gap may be even bigger if we dropped conservative atheists from the samples as they seem especially likely to suffer lower levels of well-being.
I have heard it argued that the political, and possibly religious, gap is due to differential levels of self-reporting. But as Burge shows, among the most critically ill, this gap in self-reporting is not very big. I never bought that argument as the strength of the evidence of political difference was just too much for me to accept it. But this is even more empirical confirmation that this explanation does not hold.
The data collected during Covid is very interesting. I wonder if the political and religious gaps have grown since Covid or not. I suspect they shrunk during Covid as the trauma of that time impacted all of us.
In data I have worked with the political differences are more strongly correlated to mental well-being than the religious differences. I am much more of a Christian than a political conservative so I do not say this to boost myself but I do suspect that while the religious effect matters in and of itself, the political effect is the most powerful driver of what is happening.
Political difference can be almost completely mediated by adherence to identity politics. It may be possible that this is part of the religious effect as well. One can envision a theory whereby more secular individuals are more likely to use identity politics as a pseudo-religion and thus if adherence to identity politics leads to lower levels of well-being then it is possible that secular individuals are more like to suffer for supporting identity politics than those with higher religiosity.