Scholars and analysts have much to research about the 2016 election, but some early analysis is confirming what we have seen for some time among the electorate, particularly in two key areas.

The first confirmation of past trends is what most political scientists and analysts continue to point to as the profound vote division based on party identification. Using exit poll data, barely 10 percent of self-identified partisans defected in 2016’s presidential election, continuing a strong relationship between presidential vote loyalty and partisan identification over time.

During the last two decades of the 20th century, partisans, especially Democrats, were more likely to abandon their party’s presidential candidate, particularly in the disruptive 1992 and 1996 elections with Ross Perot’s third-party candidacy.

Since 2000, however, the rate of vote loyalty from self-identified partisans to their party’s presidential candidate has been consistently within 90 percent.

Another key area that has developed over time has been the urban-rural divide, something that has become evident in North Carolina. Nationally, early analysis of this divide is showing the gulf deepening between cities and rural areas: more densely populated counties moved more Democratic, while the least dense counties moved more Republican.

In North Carolina, the urban-rural divide is really the “urban versus everyone else” divide. The state’s urban counties, which made up 54 percent of the total votes cast, split 56-44 for Obama in 2012; in 2016, they went 55-40 for Clinton.

Meanwhile, suburban counties, which make up another 20 percent of North Carolina’s total votes, went even heavier for Trump than they did for Romney. In 2012, suburban counties went 62-38 percent for Romney; in 2016, they went 64-32 percent for Trump.

Rural counties, which accounted for less than a quarter of all the votes cast in the state, went 59-37 percent Trump, widening the 57-43 split for Romney just four years ago.

In looking at how counties performed from four years ago, most of the state’s counties moved decisively to one candidate over the other.

Nine counties moved more than 5 percent in favor of Clinton from 2012’s Obama numbers, with a total of nearly 75,000 more votes than four years ago. Only one of those nine counties—Watauga—is a rural county, albeit home to Appalachian State University; the other eight are either urban or suburban counties.

However, 72 counties moved more than 5 percent away from the Democratic presidential nominee compared to Obama’s numbers, with a total loss of over 90,000 votes from four years ago.

Conversely, 62 counties moved more than 5 percent in favor of Trump from 2012’s Romney numbers, with a total of 123,000 more votes than four years ago. Forty of these counties are rural, with only four being classified as “urban” (and among the smaller of the urban counties).

Among the six counties that moved more than 5 percent away in Trump’s numbers from four years ago, five were the largest urban counties in total vote delivery: Forsyth, Guilford, Wake, Mecklenburg and Durham.

More analysis will be forthcoming, but the gap between those who identify with a party and those who live in urban areas versus elsewhere seems to be a major component of how not only the country, but North Carolina, voted. The likelihood is that these divides, among others, may continue to grow into the future.

Dr. Michael Bitzer is provost and professor of politics at Catawba College. This column is from the blog he writes for WFAE radio, The Party Line.