Granted that the state of Connecticut only has 7 votes in the Electoral College, but as with most American presidential elections, every single vote may be the difference between who takes the oath office on January 20, 2017. So while the Constitution State is not a member of my Swing State Symphony, let’s take a look what the polls from that state are saying.
In my baseline tally of the states, Connecticut is categorized as a “Solid” state for the Democratic Party. This, as with most of my baseline categorizations, is due to the history of the state in past American presidential elections. In the past seven elections (since 1988), Connecticut has voted for the presidential candidate from the Democratic Party the last six times in a row. For the last five elections running, the Democratic Party candidate has won a majority of the votes cast in Connecticut. Over the past four elections (since 2000), the median margin of victory for the Democratic Party candidate has been 17.4%. With this history as backdrop, Connecticut (and its 7 votes in the Electoral College) is solidly Democratic.
So what are the presidential preferences of the citizens of Connecticut nowadays?
The most recent polling done was published on June 7 by Quinnipiac (pronounced KWIN-uh-pe-ack) University. In that poll (1,330 registered voters, margin of error 2.7 percentage points, results here), the presidential nominee of the Democratic Party, Hillary Clinton, defeats the presidential nominee of the Republican Party, Donald Trump, by a tally of 45%-38% – or by a margin of “only” 7 percentage points.
For comparison, Barack Obama (Democratic Party) defeated Mitt Romney (Republican Party) in the 2012 presidential election by a margin of 17.3 percentage points (Obama’s 58.0% to Romney’s 40.7%).
I could only find two other polls from Connecticut taken since the 2012 election. An April 2016 poll from Emerson College (1,043 likely voters, margin of error 3 percentage points, report here) has Clinton over Trump by 8 percentage points (48%-40%). A Quinnipiac University poll from October 2015 (1,735 registered voters, margin of error 2.4 percentage points, results here) shows a Clinton victory over Trump by a margin of 7 percentgae points (47%-40%).
For further comparison, in the 2008 presidential election, Barack Obama defeated John McCain by 22 percentage points (Obama’s 60.6% to McCain’s 38.2%).
I mention the comparison to the previous pair of American presidential elections because the nominee from the Democratic Party won in 2008 and 2012 by double digits both times. Compare that to the three surveys taken since 2012 where the standard-bearer of the Democratic Party only wins by single digits (albeit high single digits).
Given that the level of support in Connecticut for the Republican Party nominees has been consistent from the previous pair of elections (40.7% for Romney, 38.2% for McCain) to the last three polls (40% (Quinnipiac October 2015), 40% (Emerson April 2016), and 38% (Quinnipiac June 2016)), can we figure out why Clinton’s level of support (47%, 48%, and 45%) has dropped below Obama’s 58% and 60.6%?
Probably not, but that doesn’t mean I won’t try?
For this comparison, I will be using the exit polling results* of the 2012 presidential election from NBC News (details here) and the latest survey from Quinnipiac.
In 2012, 63% of the female voters cast their vote for Obama (versus 36% for Romney). In the latest Quinnipiac survey, Clinton does win a majority of the female vote with 53% (versus 35% for Trump). From Obama in 2012 to Clinton in 2016, that’s a drop of 10 percentage points in the female vote (while Trump only loses a single percentage point from Romney).
For men, according to the exit polls, 51% cast their vote for Obama (versus 47% for Romney). In June of 2016, Clinton loses the male vote to Trump 36%-48%. From Obama in 2012 to Clinton in 2016, that’s a drop of 15 percentage points (while Trump gains a single percentage point from Romney).
In 2012, according the exit polls, 51% of the respondents who identified themselves as “Independent or something else” voted for Obama (versus 46% for Romney). In June of 2016, 36% of the respondents who identified themselves as Independent would vote for Clinton (versus 41% for Trump). From Obama in 2012 to Clinton in 2016, that’s a drop of 15 percentage points (while Trump drops 5 percentage points from Romney).
In November of 2012, Obama won the White vote with 51% of that demographic (versus 48% for Romney). Three and half years later, 39% of White respondents support Clinton (versus 44% for Trump). From Obama in 2012 to Clinton in 2016, that’s a drop of 12 percentage points (while Trump drops 4 percentage points from Romney).
In 2012, voters over 65 supported Obama 54%-46% over Romney. The June 2016 results from Quinnipiac show that voters over 65 support Trump over Clinton 45%-36%. From Obama in 2012 to Clinton in 2016, that’s a drop of 18 percentage points (while Trump loses a single percentage point from Romney).
Gender, party affiliation, race, and age are just part of the demographic puzzle that help to explain why Clinton’s level of support is not near Obama’s level from 2012. While it is true that results from an actual election in November are different from results from surveys released in June before an election, there is data here for each party organization to use to help identify opportunities and challenges for Election Day 2016.
Connecticut is categorized as “Solid” in my baseline tally, but that’s only due to its history. The future of the Constitution State (and its 7 votes in the Electoral College) is not yet written.
*Yes, I realize exit polling results are not the same as actual ballot results, but I work with what I have.