Census Bureau data released Thursday indicates that the percentage of foreign-born residents of the United States is at its highest level since 1910. This is not a sudden development; as we wrote last year, after sinking below 5 percent in 1970, the percentage of foreign-born residents of the United States has increased steadily. Data from Pew Research (current through 2014 at that point) shows the trend. That we’ve now matched 1910 isn’t a surprise, as such.

We addressed the subject then because comments from Stephen K. Bannon, then an adviser to President Trump, were in the news. Bannon had claimed that 20 percent of the country had immigrated here, an exaggeration of the actual figure, and that this was the “beating heart of [the] problem” of Americans being unable to find jobs.

That rhetoric is undercut sharply by the fact that this increase in immigration has, in recent years, been matched by a sharp drop in unemployment. But Bannon’s rhetoric is often less about representing reality and more about making a political argument. Over the course of the 2016 election, while running Breitbart News and then as a senior official on the Trump campaign, Bannon leveraged concerns about immigration to Trump’s political benefit, whether those concerns were warranted.

It’s worth considering, then, how the increase in the foreign-born population overlaps with politics. To assess that, we pulled data from the Census Bureau’s annual estimates by county of the nativity of residents. (As these are annual numbers, only larger counties have large enough populations for statistical significance, and, therefore, not all counties are included on the charts below.)

The first thing we notice is that there’s an overlap between how counties vote and how dense the foreign-born population is in the county.

We’ve scaled the circles on that chart to raise another differentiation. When we consider the foreign-born population, we often think largely of immigrants from Latin America. But as our analysis last year noted and as Brookings Institution analysis of the new Census Bureau data suggests, more foreign-born residents of the United States come from Asia than from Latin America. The larger the circle on the chart above, the larger the percentage of the foreign-born population in that county which is from Latin America.

That chart above, though, is not necessarily proving that foreign-born residents push counties to the left. First of all, some significant portion of foreign-born residents are not citizens, and so they can’t vote. Second, there’s another factor at play: Counties with a greater density of foreign-born residents are also places with a higher density of the population in urban areas.

Put another way, immigrants often live in cities.

And cities generally vote Democratic. They did so in 2016, as the chart below indicates. There’s a correlation between the urban population in a county and its vote.

(In that chart, we’ve rescaled the circles to represent the overall population of foreign-born individuals in the county, not just those from Latin America.)

There’s an interesting detail buried in these charts. If we consider the change in the foreign-born population, particularly the change in noncitizen foreign-born residents per county from 2008 to 2016, the places with the biggest increase in that population since 2008 were more likely to vote Republican.

This, too, is misleading: Those counties with larger increases tend to be smaller. After all, a bigger percentage increase is more likely when starting from a smaller population size.

If we scale the circles to the populations of the county, the chart looks like this:

Clearly Trump’s rhetoric on immigration, often echoing Bannon’s, was resonant in the 2016 campaign. Clearly, too, places with more foreign-born residents voted more heavily against him. But the picture that’s presented is much more complicated than it might at first appear.

As is the nature of the foreign-born population itself.