I think it all went very well - at least, I managed to get through my bits without stumbling or going blank. Hopefully some of it made sense too!
One thing I didn't make sufficiently clear is why I chose 'prayer frequency' as the indicator of religiosity. The problem is that many people - especially the educated and well off, with high social status - believe in a distant, impersonal god.
They may find their beliefs important, but it's a very different kind of belief to those who believe in a personal god who intervenes in the world.
As a result, people who believe in an impersonal god don't pray very often. In fact, the evidence from the US (the Baylor Religion Survey) is that people who call themselves atheists pray as often as those who believe in an impersonal god.
So, by measuring prayer, you're measuring a very specific kind of religious belief. Your measuring belief in a god who will not only listen to an individual, but might also take action to help them.
And that's exactly the kind of belief that you might expect people to turn to in difficult times.
This work by Tom Rees is licensed under a Creative Commons Attribution-Share Alike 2.0 UK: England & Wales License.