Interracial Marriage in the US: The Pew Numbers

The percentage that gets pasted into other people’s posts is 17%, and it is already being asked to answer questions it never took on. In 2015, Pew Research Center found that 17% of U.S. newlyweds married someone of a different race or ethnicity. The report was published in 2017. The weddings it counted were from 2015. Call 2015 the current year, or swap in the year you happen to be writing, and the sentence stops being Pew’s.

People still ask us to refresh it. A later survey would be a different document. Until that document is the one in front of us, the marriages stay in 2015 and the citation stays in 2017.

Gretchen Livingston and Anna Brown wrote it. The title is “Intermarriage in the U.S. 50 Years After Loving v. Virginia,” published 18 May 2017. The chapter these figures come from is https://www.pewresearch.org/social-trends/2017/05/18/1-trends-and-patterns-in-intermarriage/. We are not stapling a second institute onto that so the page looks fuller.

What goes wrong is rarely the arithmetic. “Who got married that year?” is not “Who, among people already married, has a spouse of another race or ethnicity?” Pew answered both. The larger answer is the one that travels.

Interracial marriage statistics

In 2015, 17% of U.S. newlyweds married someone of a different race or ethnicity, up from 3% in 1967. Both ends of that comparison are newlyweds. Neither means “all couples,” and neither means now.

The quieter finding is the one we would rather you use for the ordinary question. In 2015, 11 million married people were intermarried, 10% of all married people. If someone asks how common interracial marriage is among people who are married, say 10% and keep the year. Use 17% when the subject really is people who married in 2015. If that cohort is your subject, ignore the preference. It follows the question, not a taste for smaller numbers.

Eleven million counts people, not couples. Leave it there. Splitting it into a number of marriages would be a calculation these findings do not give us, and we are not going to improvise one.

Pew’s phrase on the 17% line is “a different race or ethnicity.” That bucket is wider than a black spouse and a white spouse. A reader who only means those two groups cannot get the answer out of 17%. The share will look too large, because other pairings are inside it. Livingston and Brown published more rows than the four findings on this page. If you need another group’s rate, read the chapter. Supplying the rest from memory is how stray digits show up next to a real source.

Among black newlyweds in 2015, 24% of men and 12% of women had a spouse of a different race or ethnicity. Each percentage sits on black newlyweds of that sex, not on black adults and not on black singles. Someone who did not marry in that window is outside both figures. So is any story about why. The table records spouses. It does not record motives.

This is where a count gets drafted into a theory of dating. Twenty-four and twelve do not show who is more open, who wants marriage more, or who has an easier time on an app. We don’t accept that translation. If you have evidence about desire, or about apps, bring it and label it as that. This report is a count of spouses among people who got married in 2015. You can disagree with the refusal. The disagreement still needs a source that is not these two shares.

Couples with one black spouse and one white spouse were 11% of intermarried couples. About 7% were a black husband and a white wife. About 3% were a black wife and a white husband. In the chapter, that mix describes intermarried newlywed couples in 2015, not every marriage already underway. The “about” is Pew’s. Seven and three do not add to eleven. Rounding is allowed to look messy. Forcing the slices to add is how a clean chart starts editing the report.

Set next to the newlywed split, those couple shares describe a pattern and stop. In these 2015 counts, a black husband with a white wife was the larger of the two black-white arrangements, and black men who were newlyweds more often had a spouse of another race or ethnicity than black women who were newlyweds. The cells do not rank anyone as a partner, and they do not explain themselves. Whatever account you have already heard came from somewhere else. Ask where, before you repeat it under Pew’s name.

Eleven percent is not most. Photographs beside pieces like this one still tend to show one black partner and one white partner, because that is the couple the headline promised. The photo is narrower than the 17%. The 11% is the figure that matches the photo. For a site aimed at black and white dating, that is slightly awkward, and it is still the finding. Using 17% as a headcount of couples who look like the photo is the mistake we most want to retire.

Loving v. Virginia is the 1967 case that ended state bans on interracial marriage, and 1967 is the year of the 3% newlywed figure, which is why the case is in the title. We can name the case and what it did. We are not adding a vote count, and we are not quoting the opinion. A line reconstructed from memory does not belong next to numbers this specific. State bans had ended. The newlywed share was 3% in 1967 and 17% in 2015. A causal story with more parts than that has to come from a source that argues it.

What drops out of the sentence

A citation you can check is dull on purpose. Name Livingston and Brown, say the report was published on 18 May 2017, say the figures are for 2015, and keep the group in the sentence. “17% of U.S. newlyweds in 2015” still means what Pew measured. “17% of American marriages” does not. A version that says the share applies now does not either.

Eleven percent of intermarried newlywed couples will not tell you if yours is the only black-white marriage on your block, or if the two families will sit through a meal. A national mix is a bad map of one household. These findings also say nothing about how long the marriages last, whether relatives approved, or what strangers said.

That part is on the interracial couples page, without this percentage stack copied over. If you are not at marriage yet, the black white dating app page is a product description. Don’t look there for a revision of Livingston and Brown, and don’t look here for a price or a star rating.

Who married whom in 2015 is not the same fact as who told a pollster they approved. The report has a chapter on public views, and we are leaving it alone. An approval percentage, from Pew or from anyone else, would mix two questions and pretend they were one. If that is what you came for, you want a page about that survey. You do not want it borrowed.

Before you copy a sentence from here, check that a stranger could tell whether you mean 2015 newlyweds or all married people that year, and whether you mean any intermarriage or a black-white couple. If either detail dropped out, the sentence is already wrong, even with 17 still sitting in it.

The report is old enough that swapping in a fresher year is tempting. It is still a misquote. If you need the 2020s, get a newer release from the people who fielded the survey. Leaving this one dated is the accurate way to use it.