In an effort to better understand the 2016 electorate, we matched members of our nationally representative American Trends Panel to a national voter file to create a dataset of verified voters. (Thread on findings follows.) pewrsr.ch/2KFmuSe
Among these verified voters, the overall vote preference mirrors the election results very closely: 48% reported voting for Hillary Clinton and 45% for Donald Trump. (Actual election result was 48% Clinton - 46% Trump)
This data source offers a detailed look at the voting preferences of Americans across a range of demographic traits and characteristics. 
This data also allows us to look at how the 2016 electorate voted by party and ideology. Voters in 2016 were deeply divided along ideological lines.
We can also examine the 2016 vote by religious affiliation. As in previous elections, voters in 2016 were sharply divided along religious lines.
One unique insight our data offers is a look at the demographic composition of the coalitions of voters who supported Clinton and Trump.
As this chart shows, Trump voters were more likely than Clinton voters to be rural and white, and less likely to have attended college. Clinton voters were more likely than Trump voters to be urban, racial or ethnic minorities and college graduates.
Finally, this data also provides a way for us to compare 2016 validated voters with nonvoters in our American Trends Panel.
As this chart shows, nonvoters were more likely than voters to be younger, less educated, less affluent and nonwhite. And nonvoters were much more Democratic.
You can find our full report on the 2016 electorate here: pewrsr.ch/2KFmuSe
For a summary of findings on views of President Trump among those who voted for him in the 2016 election, see this Twitter thread:
For a summary of findings on views of President Trump among those who voted for him in the 2016 election, see this Twitter thread:
The “share of electorate” column in this chart has been edited to reflect updated percentages for gender by race to correct for a data tabulation error. Changes do not affect the report’s substantive findings. The correct chart is attached.
