Deconstructing the White Vote

In this post, I am going to try and show that the White vote is more heterogeneous than often thought, or, at least, a lot more heterogeneous than I used to think it was. If you look at Whites as a whole, we look to be pretty consistent republicans:


In a previous post, I documented that immigrant White Americans are, on net, Democrats. I showed this first by analyzing party ID:


I then looked at the voting patterns of 2nd generation White Americans:


Finally, I looked at the party ID of 3rd generation White Americans, sorted by the number of grandparents they had who were born outside of America:


More recently, I ran some similar analyses of White Americans by ethnicity. I’ll be relying on data from the General Social Survey, and utilizing a question that asked participants ““From what countries or part of the world did your ancestors come?” Here is the main result of the analysis:


Thus, only British, German, and Scandinavian Whites were found to identify as Republicans.

To ensure that this isn’t somehow an artifact of people being mistaken about their ancestry, I calculated these same ratios for people who were either 2nd or 3rd generation immigrants to the US. People who have at least one parent or grandparent who were immigrants should have a much more accurate knowledge of where their family is from. The Spearman rank order correlations between the ratios produced by the total sample and 2nd and 3rd generation immigrants were .79 (p<.001) and .89 (p<.001), suggesting that whatever misremembering does occur in the total sample does not bias it in a way that messes with the pattern of political ideology by European ethnicity.

I also tried looking at why European ethnicities differed in party ID. The only trend of value I found was that people from more protestant nations were more Republican. This trend explained about a third of the variation across ethnicities in party ID:


I went on to analyze voting trends for the 12 US presidential elections that occurred between 1968 and 2012. Due to limitations of sample size, I only looked at people from England, Scottland, France, Ireland, Italy, Poland, and Scandinavia (Denmark, Noway, and Sweden, combined).


If you want, you can see each ethnic groups voting pattern by election below:




A common pattern across these is that Whites liked Reagan a lot, more than they liked most Republicans, and liked Clinton a lot, more than they liked other Democrats. There is an interesting divergence in the Obama years wherein the Irish, Italians, and Scandinavians, flipped to the Democrats.

So, that’s what I found. Let’s look at what others come up with on recent voting among European American ethnic groups.

BuzzFeed reported on a poll that looked at party ID by the ancestry that White people most identified with. This forced respondents to chose only one ethnicity, and this is potentially problematic given the nature of the American population and the fact that some ethnicities enforce identity more than others. Most people are descended from multiple nations, and so asking the question like this allows them to chose from multiple possibilities, and a dubious “American” category, that could mess with the results.

Voters were asked if they would support a Republican or Democrat, if a congressional election were held in their district today. Groups and margins of error as above.

Clearly, their sample bound Irish and Italian respondents to be more to the right than mine did, thought the Irish were still found to be net democrats.

This poll was carried out a few months before the election of 2016 and found that respondents of each ethnicity intended on voting for Trump.

The poll included 429 white voters who identified most strongly with German ancestry, 533 English, 344 Irish, 211 Italian, and 911 American. Margins of error run from +/- 3.25 percentage points for American to +/- 6.75 percentage points for Italian.

Now, this finding I don’t put much stock in. If accurate, this would imply that 20% of the English, Italians, and Irish, voted for a third party, which they obviously didn’t. This is why asking people after the fact is better, they have actually had to make the choice rather than sit in a place of indecision.

There are a few other data points on how the Irish voted in 2016. A poll reported on by The Irish Times found that 47% of Irish Americans voted for Clinton and 27% voted for Trump. The same survey found that 41% identified as Democrats and 21% as Republicans. However, this poll was biased in that its respondents were disproportionally from the North East, possibly more so than is appropriate for a poll of Irish Americans.

Another analysis looked at the 14 counties in which the Irish comprise at least 25% of the population. Clinton won 8 of them. This means that Clinton won 57% of Americas most Irish counties while she only won 15.5% of US counties generally.

With respect to Italians, there is less data. My analysis ended in 2012 and found them to Democrats at that time. This is supported by a 2011 Zogsby poll which found that 37% of Italian Americans identify as Democrats while 30% identify as Republican and 31% identify as independent. It is possible that BuzzFeed measured a real change in Italian opinion. This wouldn’t be surprising given how well Trump did in Pennsylvania compared to previous Republicans.

Buzzfeed didn’t look at Polish Americans. Radzilokski and Stecula (2010) find that Polish Americans identify as Republicans 26.1% of the time, Democrats 36.5% of the time, and independents 33.2% of the time. They also identify as conservative 44% of the time and liberal 33% of the time, suggesting they may have a lot of conservative Democrats.

My analysis began with the year 1968. Unsurprisingly, earlier data is consistent with the idea of various White minorities voting Democrat.  For instance, Wolfinger (1965) finds that Italian and Irish in the North East were democrats in 1958. Craughwell (2012) reports that “JFK won 75 percent of the Irish vote, 75 percent of the Italian vote, 82 percent of the Slavic vote and 85 percent of the Hispanic vote.”

So,  it looks to me like there are groups of White people that aren’t reliable Republicans, especially in the last decade, and that the political assimilation of White ethnicities have been exaggerated. Of course, in a sense, we already that there were Whites who are not reliable republicans since lots of White people are Democrats, but I previously didn’t appreciate the degree to which this depended on how long White people had been in the country and where they came from in Europe.

3 thoughts on “Deconstructing the White Vote

  1. Pingback: The Myth of European Assimilation | Ideas and Data

  2. Pingback: On White Nationalism | Ideas and Data

  3. Pingback: The Leftist Influence of Catholicism | Ideas and Data

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s