The quiz seems to be carefully tuned to detect potential inconsistencies that stem from a belief that a god must be omnipotent. Though nowadays I'm an atheist materialist, I was raised as a Hindu, so I don't have any emotional problem with finitely-potent gods.

Anyway, I think Dawkins and Dennet are full of crap when they talk about memetics.

Rational-choice theory suggests that there are plenty of fully rational reasons (in the full no-holds-barred intertemporal utility maximization homo economicus sense) for people to join religions and observe weird ritual restrictions on their behavior. The basic argument runs like this: a religous group provides an altruistic social network for its members, and as such is subject to a free-rider problem, as people show up to collect the benefits but don't contribute very much. So this is where complex ritual limits on behavior show up: they make it easier to identify free riders. If you have to observe complex rituals, then it's easier to distinguish the genuinely committed from the free riders.

Also, establishment of an official religion tends to reduce participation, because a) a monopoly makes it harder for individuals to find a religious variant that appeals to them, and b) priests in a monopoly church have a reduced incentive to make their religion appealing. (It was on the strength of this last argument that Adam Smith opposed establishment and David Hume supported it.)

So has a number of implications. First, that stricter religions will grow more quickly than more lenient religions, up to the point where the strictness becomes so expensive it overwhelms the benefits of the religion. Second, as you go higher up the socioeconmic ladder, the less strict popular religions will become, since the opportunity cost of pulling away from the rest of society rises.

All of these predictions seem to hold true in practice within the US and Europe. Stricter variants -- such as Southern Baptism, Mormonism and evangelical Christianity -- have been growing much faster than more liberal churches like the Episcopalians. They also tend to have less well-educated and affluent members than the liberal churches. Finally, religious participation rates are much higher in the US than in Europe, which has never had any established religion, and the number of sects is also substantially greater. (One exception is that these results don't hold nearly as strongly for Catholicism as for Protestant sects: to explain the difference one must presumably look at differences in institutional structure.)

I like this line of research because it presupposes that people are smart, and choose to do what they do for good reasons. Forward-looking, intentional behavior is real and significant, and I don't think that the memetic drive to define it out of existence is at all realistic. In the late 50s Noam Chomsky showed that you can't model linguistic behavior with Markov models, and basically all the standard models of evolutionary game theory can be represented as Markov models. So I'm extremely dubious about the utility of memetics. Well, except for fiction. It seems like you can build more plots around the idea of viral ideas than around everyone acting foresightedly. :)