"Only 6% of Scientists Identify as Republicans." I've run into that statement countless times on USMB and it surely didn't jive with personal experience in Silicon Valley and elsewhere.. So I started to look into the claim.. Found one USMB thread from last year that ended up in the "tantrum room". Don't want THAT to happen here. What I what to do is nail the original source of that oft-quoted meme and do a factual analysis of the claim... So far -- this is what I know.. Origin of the "factoid" is a Pew Poll that was done in conjuction with American Association for the Advancement of Science (AAAS), a generally respectable organization that publishes "Science" magazine. Membership is similiar to National Geographic Society and all that is required is a subscription and a questionnaire. The 6% factoid was only ONE of a couple dozen questions posed EXCLUSIVELY to AAAS membership.. But that was ALL that the political rags like HuffPost and Slate needed to know to turn it into flaming partisian dung. The actual "factoid" can be found on this page of the survey.. Section 4: Scientists, Politics and Religion | Pew Research Center for the People and the Press So if AAAS is a truly respectable organization and is composed of "scientists" --- what could be the problem with the claim that only "6% of Scientists Identify as Republicans"? 1) We would need to know if this is the ONLY POLL ever done on the subject and whether there is further evidence of this fact ---- I can't find a single assertion of fact on this question that does not refer back to the Pew/AAAS poll cited here. 2) Is there any potential institutional bias due to the organizations involved? --- Both are highly respected organizations, tho leaning somewhat to the left. PEW's employees contribute almost exclusively to Democrat candidates and causes. And AAAS has many "social justice" initiatives that might seem more lefty political than scientific. 3) Does the statistical sampling of the AAAS membership ACTUALLY REPRESENT the composition of the general population of "scientists"? -- Most probably not.. And I'll elaborate on that below... Question #3 presents the biggest problem for this factoid. I am certain that the sample population for this SINGLE poll does not accurately represent the general population of "scientists" in America stating political preferences. And any PROOF of this factoid would require far more extensive polling.. The Pew sampling methodology is located at: http://people-press.org/files/legacy-pdf/528.pdf and gives some breakdown on the AAAS sample... To wit:: I seriously doubt that the general composition of scientists is anywhere near 63% academic and only 15% industry. Even if it were true -- that alone would partially explain the result. And with the last survey question adding up to WAAAY more than 100% , there's some question as to the allowed answer(s) for whether their work was largely Teaching and administration or real research and clinical practice (or some combo of both if in a university). Perhaps others want to contribute SCIENTIFIC analysis to this question in the form of how the general "scientist" population breaks down between academic and industry. Keep in mind that NO PARTICULAR effort was placed in verifying the bona fides of the individuals surveyed, no control was placed on geographic regions.. I find it very likely that this factoid is merely an opportunistic political use of a single data point that was NOT derived in a very scientific manner. Truly an irony in itself.. DO NOT CONTRIBUTE TO PUTTING THIS THREAD into the tantrum room.. I tried hard to be objective and non-political in the OP construct.