One of the striking results from my survey on natural theological arguments is that most philosophers of religion are theists. Even if I restrict my count to a subsample consisting only of those people who are philosophers, who have listed philosophy of religion as one of their areas of specialization, and who are faculty or non-faculty with PhDs, the sample is overwhelmingly theist. Of this select subsample (N = 118), 70.3 % are theists, 16.9% atheists and 12.7% agnostics (the rounding explains why we are not at exactly 100 %). As you may recall, the percentage of theists slightly higher (around 73%) in my general sample philosophers of religion, which also includes graduate students, undergraduates and those outside of academia. Given that the PhilPaper survey gave a similar result, we can be highly confident that about 7 in 10 philosophers of religion are theists. One of the discussions of my preliminary results on Prosblogion is whether we should accord any evidential weight to this (i.e., should we defer to the expertise of those who are studying the existence of God), or whether this should lead us to an increased skepticism about philosophy of religion as a discipline.
There is a lively debate in social epistemology about under what conditions we should adjust our credences in the light of the conflicting opinion of epistemic peers. Conciliationists like Christensen and Feldman argue we should at least revise our confidence downward, giving equal weight to those who are our epistemic peers. Those who endorse a steadfast position argue that we can stick to our guns. There are many variations on this general theme, but it seems there is at least something of a consensus that we should bring our beliefs in line with our epistemic superiors. This does not mean that we should completely defer to the expert's opinion. After all, as Elga already recognized, it is only in highly idealized circumstances reasonable to defer to someone's opinion no matter what that opinion is. One can reasonably discount information that falls outside the appropriate range (e.g., the plumber informing me that my heating does not work because it is haunted by ghosts). But it seems appropriate to defer to a plumber who says the heating is broken because of some malfunctioning part he points out to me. Moreover, Alvin Goldman writes: "to the extent that it is feasible, N should consult the numbers, or degree of consensus, among all relevant (putative)experts. Won't N be fully justified in trusting E1 (expert 1) over E2 if almost all other experts on the subject agree with E, or if even a preponderance of the other experts agree with E?"
I am not advocating any close similarities between philosophy and normal science. But supposing that philosophers do have some form of expertise and knowledge about their domain of inquiry, we could use Goldman's principle of deference to expertise in the following way: it is rational/reasonable to adjust my beliefs in accordance to what the majority in a particular AOS believe. For example, suppose I earnestly believe that taking the one box is the correct solution in Newcombe's problem. But I then learn that decision theorists resolutely choose the two boxes. It seems plausible I should revise my confidence in my original belief downward.
Now consider philosophy of religion. Prima facie, if we applied Goldman's principle this seems the road to take. However, if self-selection and confirmation entirely explain the prevalence of theism in PoR, this is not the right course to take. After all, in consulting astrologers, who are the supposed experts in astrology, I would find that the large majority of them believe that astrology works. As Alexander Pruss argued in the comment section here, if God exists, philosophy of religion is one of the most important areas of philosophy, but if He does not, it's peripheral. So we can expect a self-selection effect. Combined with confirmation bias, there is an additional worry that theistic philosophers of religion might overvalue the strength of theistic arguments, since theists in general evaluate these arguments as stronger. In fact, I found no significance for PoR as an AOS in how natural theological arguments were evaluated taking religious belief into account. Confirmation bias does not disappear even in people with high levels of education. As one commenter on my original entry suggested "My reading of the situation is that philosophy of religion is unhealthy, and further that your data are best understood as demonstrating a statistically significant conflict of interest."
If a self-selection effect is the sole drive of theists to the field of philosophy of religion, and confirmation bias subsequently plays a role in their evaluation of theist arguments, we should be perhaps wiser to suspend judgment. The effects of confirmation bias on the rationality of our beliefs is a bit too large in scope to treat here. Suffice it so say that I am a Kuhnian at heart; I do not worry much about confirmation bias, given its prevalence it makes more sense to counter it by improving diversity within a profession. As I have argued in several papers (e.g., here), a discipline's health does depend on having multiple, disagreeing voices. So philosophy of religion could benefit from having more atheists and agnostics.