In imitation of Steve Sailer's--epigone, after all!--detailed demographic breakdown of the 2012 presidential election with Romney and Obama going head-to-head, here's a less aesthetically appealing version of the same for 2016 with Trump and Clinton squaring off.
A few brief technical notes: Reuters-Ipsos never released official exit poll results, but they have maintained a daily tracking poll on Trump's approval rating with filters including who respondents voted for in addition to all the various demographic categories presented below. The data is taken from the poll's commencement on January 20 through March 14, the latest date for which figures were available at the time of this post's creation. The total sample size among those who voted for either Trump or Clinton is 13,381*. Trump shows a modest edge, 50.6%-49.4%, in the popular vote in this sample. I have no opinion on whether this is more or less accurate than the officially reported popular vote total when it comes to legitimate ballots cast by American citizens.
The following graphs show the vote share of each demographic group Trump received in November:
Married women were considerably more likely--by a margin of 9 points--to vote for Trump than unmarried men were. They were even marginally more likely to vote for him than married men were! While only 3% of unmarried black women voted for Trump, more than 1-in-4 married black men did (n = 284). Married Hispanics were marginally more likely to vote for Trump than unmarried whites were. Marrieds were more likely to support Trump than singles in every category examined.
While Mormons didn't come out as overwhelmingly for Trump as they did for Romney, most of the latter's co-religionists did not reveal themselves to be made of the same quisling cuckery that the GOP's 2012 nominee is made of.
I inadvertently didn't include all Jews as a single category in the visual representations, but 35.2% of them went for Trump (n = 378). That's a significantly better performance than the 24.5% figure the officially commissioned exit poll reported.
I also failed to include the non-negligible chunk of the electorate without any religious affiliation (n = 2,119). Trump got 31.9% of the heathen vote.
Among those of family formation age, Trump wins by 6 points among those with kids while losing among those without kids by 12 points. This isn't just among whites, it's among all voters. Additionally, it excludes those aged 50 and over (who lean Trump). So sans immigration, Trump gains and Clinton loses as time goes on.
If there are other demographic subgroups of interest, let me know in the comments and I'll take a look at them (and eventually revamp the graphs to include those additions).
* Sample sizes for the following demographic groups numbered less than 200: Single Jewish men (39), single Jewish women (51), Hindus (53), Muslims (67), single other races (90), Married Jewish men (102), Black single mothers (104), married Jewish women (119), married Hispanics (157), married other races (165), other sexual orientation (166), single Asians (168), and single Hispanics (192).