Update: Just to be clear, I think the variation across cultures is probably explained in large part by confusion as to what is being asked, and differential sampling. In particular, I suspect that the 'Turkey" sample is more representative than the "Bangladesh" sample, because Turkey is a more developed society. I've mentioned before that many (most?) Muslims are Creationists, broadly understood. According to Pew's Religious Landscape Survey 42 percent of American Muslims accept that evolution is the best explanation for the origin of human life on earth. This is roughly in line with the American public, if a touch on the Creationist side. The numbers are similar in Turkey. Also, it must be mentioned that unlike most I have some experience with educated (and scientifically trained) Muslims, and can attest to the fact that many are Creationists (my family). So the results of a new survey of the world's Muslims by Pew took me aback a bit, in that it reports widespread acceptance of evolution among Muslims. To add to the plausibility the results for Turkey are in line with previous findings: a bit more of Turkey's population are Creationist than not. The results for highly secularized European Muslim populations are plausible, though the gap between Albania and Kosovo is somewhat strange. But look at the results for Bangladesh and Lebanon! I have to admit some skepticism. My concerns are twofold: first, many of these questions may be interpreted differently from society to society, so that comparison may be difficult. This is why I tended to focus on within-region comparisons when ingesting the other survey responses (Pakistan vs. Bangladesh, Lebanon vs. Palestinian territories). Second, I am not sure as to the representativeness of the sample. Do the opinions surveyed actually reflect the broader society? In extremely poor nations like Bangladesh I have difficulty even comprehending how illiterate subsistence farmers would interpret some of these questions, their perceptions of modern abstractions of nationality and identity are generally so inchoate. There's also a broader dynamic which needs to be addressed: modernization in many cases leads to greater 'conservatism' of belief and practice. Older subsistence farming societies are often tolerant and accepting of diversity of opinion on a macro-social scale because they are fragmented enough that such variation can be accommodated without too much controversy. In contrast, urbanizing societies characterized by upwardly mobile middle classes living cheek by jowl often exhibit simultaneous patterns of secularization and radicalization, with the latter often defined by appeals to a reversion to tradition and proper adherence to formality and ritual (often these are novel constructions and modern interpretations of ancient motifs). Turkey's Creationism in relation to Bangladesh may simply be due to the relative social advancement of the former in relation to the latter, where broad based mass popular culture has attained a level of power and self-determination to challenge elite narratives. Ultimately the terminal state of this challenge seems to be capitulation and co-option by the elites, but until that moment one is confronted by the reality of dramatic ideological tensions between the elite and aspirant elite factions.