That’s the number. At least according to Public Policy Polling. That seems rather high. So I decided to go back and look at the RACMAR variable in the General Social Survey. Here’s the question:
Do you think there should be laws against marriages between African-Americans and whites?
They kept asking the question for 30 years, but dropped it in 2002. Here’s the reason:
By 2002 it was a consistent finding that less than 10% of Americans would accede to the proposition that interracial marriage should be legally banned. So the finding that that 46% of Mississippi Republicans agree with that position, and that only 40% reject it outright, is somewhat curious. Here’s the question in PPP:
Do you think interracial marriage should be legal or illegal?
The outcomes were:
- 40% said legal
- 46% said illegal
- 14% were not sure
Remember that the sample was limited to Mississippi Republicans. Let’s go back and look at some of the demographic correlates for the responses to RACMAR between 1998 and 2002, when the proportion responding yes and no was relatively constant. I’ll focus on region and politics.
| Ban interracial marriage | |
| Liberal | 7 |
| Moderate | 10 |
| Conservative | 12 |
| Democrat | 10 |
| Independent | 10 |
| Republican | 11 |
| New England | 7 |
| Middle Atlantic | 8 |
| East North Central | 9 |
| West North Central | 9 |
| South Atlantic | 11 |
| East South Central | 26 |
| West South Central | 15 |
| Mountain | 7 |
| Pacific | 5 |
| East South Central Only | |
| Ban interracial marriage | |
| Liberal | 16 |
| Moderate | 28 |
| Conservative | 30 |
| Democrat | 19 |
| Independent | 28 |
| Republican | 34 |
The “East South Central” is a Census division. It consists of Alabama, Kentucky, Mississippi, and Tennessee. Of these states Mississippi is probably the most conservative one, so we can assume that its proportion was higher than 34 percent. Also, it is the least populous of the states, with less than half the population of Tennessee, so its weight may have been lower. It still seems to me that 10 years on from 2002 we should see a lower number than 46 percent for Mississippi Republicans, but the PPP value isn’t implausible on the face of it. Additionally, the election of a biracial liberal Democrat as President of the United States may have made the race issue more salient for white Southern Republicans.
Finally, there were two crosstabs which I thought were kind of strange:
I think the liberal Republican opposition to interracial marriage in Mississippi is due to small sample size and mistaken responses. I really doubt there are very many liberal Republicans in Mississippi. But the age results confuse me. The GSS doesn’t indicate any increased opposition to interracial marriage among conservative/Republican respondents 10 years ago in the East South Central region. On the contrary.
Razib Khan’s degrees are in biochemistry and biology. He has blogged about genetics since 2002, previously worked in software development, is an Unz Foundation Junior Fellow and lives in the western US. He loves habaneros.
April 7th, 2011 at 9:50 pm
Can you get the number of people in each category out?
I am with you, for some of these it seems likely that they are meaningless due to small numbers of responses.
April 7th, 2011 at 10:07 pm
This just makes me sigh. It surprises me that 55 percent of very liberal people are in favor of banning interracial marriage ( if the enlarged graph is the same as the one above it). That’s more than any other surveyed demographic. Hmmmmm…..
April 7th, 2011 at 10:22 pm
The GSS doesn’t indicate any increased opposition to interracial marriage among conservative/Republican respondents 10 years ago in the East South Central region. On the contrary.
Well, 10 years ago, many of the respondents who now fall into the 30 to 45 age bracket were in the youngest age bracket whereas many in the lowest age bracket were too young to be surveyed. This new resurgence of what we northerners would consider an anachronistic viewpoint may be fueled by those in their late teens and early twenties. It’s weird, but I don’t see any conflict in the data.
April 7th, 2011 at 10:26 pm
This, as people have noted, demonstrates why marriage laws should be left not to legislatures or referenda but to courts. This, one might go on, demonstrates why Mississippi shouldn’t be a self-governing polity. Disgusting.
April 7th, 2011 at 10:27 pm
#3, yeah, i thought of that. it’s not logically a faulty argument. but i’d like to see more confirmation, it seems too weird. even the noted exception to the rule re: social conservative, more pro-life youngsters vs. gen-x and baby-boomers, is a modest trend. alex, i couldn’t find the numbers. but i’m assuming that for the total republican pool the N has to be OK. but i doubt the “liberal republican” N is big at all. i think a lot of them are just confused people.
April 7th, 2011 at 10:53 pm
This, one might go on, demonstrates why Mississippi shouldn’t be a self-governing polity
fair enough. but do note that consistently applying this sort of logic (i.e., your norms of basic minimum human rights) universally will naturally lead to the dispossession of the popular voices of most of the world. since i’m skeptical of democracy as americans go i don’t hold that against you
April 7th, 2011 at 11:02 pm
Some of the crosstabs are odd.
Overall it could be plausible. This is a survey of 400 usual Mississippi Republican primary voters. Now, that is very likely a fairly stubborn sort of people who regularly vote in primaries. They could have very different beliefs and characteristics from the general population. Also, remember that the 18-29 crowd almost completely abstains from primary voting. So that could explain the weird crosstab there.
April 7th, 2011 at 11:34 pm
#7, i think you zeroed in on what the likely issue with the youth is, selection biasing of voting at that age. i was wondering if there was any reason that very young mississippi republicans could be weird and more ideological. that would explain it.
April 8th, 2011 at 1:33 am
Point, Razib, point. It’s just that Mississippi has to stand out spectacularly even within its census region. What is up with that state?
April 8th, 2011 at 2:39 am
Elvis Presley was born there. Not that I think that means much, in his teen years he was a big fan of Sister Rosetta Tharpe, who was arguably the first female rock guitarist.
Dig this: http://www.youtube.com/watch?v=JeaBNAXfHfQ
Just out of curiosity, would the same attitudes hold for mixed white/asian marriage as for white/black marriage? Just askin’. I have no idea, but imagine the % against might be lower but not zero.
I have had white British professional colleagues who went out of their way to inform me in all seriousness that they did not “approve” of my mixed marriage. But then I had white Australian professional colleagues who said even more repugnant things. I had a very repugnant answer for them.
April 8th, 2011 at 2:58 am
It’s just that Mississippi has to stand out spectacularly even within its census region. What is up with that state?
i think it’s the most uniformly ‘old dixie’ state
http://en.wikipedia.org/wiki/Appalachia
louisiana has a lot of cajuns. florida is very diverse. alabama has more appalachia. etc. mississippi is dixie qua dixie.
April 8th, 2011 at 4:35 am
The answer is obvious: they thought the question was about same-sex marriage! (Except it still makes the cross-tabs very strange.)
April 8th, 2011 at 4:39 am
Having now looked at the poll, I have alternative explanation. The question actually asked was about same-sex marriage, and the report contains a repeated typo
And the weird cross-tabs are results of small sizes in the liberal boxes, as folk has suggested.
April 8th, 2011 at 6:27 am
Republicans in Mississippi who call themselves “very liberal” are outliers in a lot of ways, I would guess. (To put it differently, people who think that Obama is a Communist probably thinkt that, for example, (ex-) Senator John Warner of Virginia was very liberal.)
In 2000 in Alabama there was a vote to repeal an old anti-miscegenation law. It passed 60-40. So there was still sentiment for keeping miscegenation legal.
I did some checking to see who voted which way, comparing the vote to the Bush-Gore vote. The general results were that the blackest counties voted for the bill (against the law) pretty strongly and for Gore, the relatively urban areas voted for the bill and for Bush, and the more rural white areas voted against the bill. The all-white area in the north voted strongly against the bill (for the law) and were more likely than white people elsewhere to vote for Gore. Roughly speaking, there was a black-white difference, and an urban-rural difference, but as for Alabama white Democrats and Alabama white Republicans, no clear difference appeared, but with a hint that white rural Democrats were at least as racist as white rural Republicans.
In Alabama as in many other states, the state Democratic Party is far out of synch with the national Democratic Party and represents local interests, and a lot of people split their tickets between federal and state. Many Southern legislatures are still controlled by Democrats, even though none of these states would ever vore for a Democrat for President.
Someone with access to the right data could easily do this comparison in a less haphazard way. I was working with inconveniently organized sites on the internet which made it hard for me to present the material.
Here’s the Alabama vote: http://www.class.uidaho.edu/engl_258/Lecture%20Notes/american_antimiscegenation.htm
South Carolina also seems to be a pure Dixie state.
April 8th, 2011 at 8:16 am
The results are not surprising considering PPP is a known left-leaning organization. Lies, damned lies, and statistics…
I’m immediately skeptical of any figures presented as such by any politically affiliated organization. It goes for all sides of the political spectrum on a variety of issues – they just use their “data” to propagandize in favor of their own party. If they were to poll Southern Democrats and got similar results, as noted by John Emerson, they simply would not make it public or spin it in some fashion to draw distance between “those” Democrats (they’d probably use “conservatives” instead to blur the distinction since it can apply to either party) and “the” Democratic party. Nobody wants “their” party to be affiliated with such negative publicity.
April 8th, 2011 at 10:12 am
I agree that these results are bizarre. At the same time, I think it’s reasonable to say the patterns among Republican primary voters (in terms of change over time) will be significantly different than the population as a whole. I just made a few graphs from the GSS data, and when I include all respondents by year, I get the same graph posted here. But if I break down the proportion supporting interracial marriage by political ideology, there is a lot more variation in the trend over years. For example, among all respondents you see only a slight uptick in opposition to interracial marriage around 1980, but if you restrict the graph to people who self identified as conservative, the change is much more drastic. While it seems that the results of this poll are unreliable, I think it does raise an interesting question. Data seem to indicate that for the general population support for interracial marriage is primarily a function of time. When restricting the analysis to conservatives/Republicans, other factors seem to play a more significant role. Perhaps certain social events drive support for interracial marriage among certain groups. In the early 1980s, there was a lot of talk of “welfare queens” that may have temporarily driven up racial resentment among conservatives (since the argument was made primarily among conservatives). Perhaps we are seeing a similar effect today.
April 8th, 2011 at 10:39 am
It’s amazing how many people fell for this planted “news” story and this nonsense poll. Public Policy Polling was founded by Democratic pollster Dean Debnam. That should be more than enough to give everyone pause. This is nothing more than race-baiting and a scam to get more votes for Obama.
April 8th, 2011 at 10:52 am
re: left leaning. THAT’S WHY I DOUBLE CHECKED THE NUMBERS FOR PLAUSIBILITY WITH THE GSS. you guys bother to read my post? please don’t view the poll in isolation when i took time to see if it was clearly a total outlier. i’m not sure it’s legit, even if there wasn’t a coding error people can be primed easily, and polls have an element of randomness which isn’t included in the margin of error. but the GSS is not a left-leaning outfit.
April 8th, 2011 at 12:31 pm
It is also worth observing that Mississippi, despite having one of the lowest percentage of non-Hispanic whites in the nation, has an almost entirely non-Hispanic white Republican party that regularly secures upwards of 90% of the white vote.
No state has more racially polarized voting patterns than Mississippi, so it isn’t too surprising that Republican disapproval of interracial marriage is so intense there.
The percentages of Republicans that favor banning interracial marriage in Mississippi is on the same order as the percentage of Republicans there who are birthers, or believe that President Obama is a Muslim, or favor secession from the United States of American right now.
Keep in mind that 46% of Republicans is probably something on the order of 20% of the state population. I suspect that the percentage of unaffiliated (but probably Republican leaning) and Democratic voters who favor banning interracial marriage is much, much lower, probably under 10%, which would be about 26% for the state as a whole, in line with GSS data.
April 8th, 2011 at 12:45 pm
Ideology raw numbers:
Would you describe yourself as very liberal,
somewhat liberal, moderate, somewhat
conservative, or very conservative?
Very liberal ……………………………………………… 4% (16) so 55% is about 9
Somewhat liberal …………………………………….. 4% (16) so 71% is about 11
Moderate………………………………………………… 17% (68)
Somewhat conservative……………………………. 36% (144)
Very conservative ……………………………………. 40% (160)
April 8th, 2011 at 1:25 pm
What I reported was that, in a formal legal vote in a state neighboring Mississippi, 40% of the voters supported laws against intermarriage. I also reported that as far as I could tell, rural whites of both parties were the main supporters of these laws. I also pointed out that the research I did was highly suggestive but not well done enough to report as fact, and I invited someone else to do the necessary, not very difficult work. All it would require would be to plug the county-by-county votes on the law and compare them to the racial compositions of the counties and the counties’ D/ R voting pattern to get interesting results.
However, by and large what I wrote makes it seem quite plausible that 46% of the Mississippi Republican party wanted miscegenation to be illegal.
There is also evidence that Palin voters are especially racist. When Barbour is not a candidate, Palin gets a significantly greater share of the miscegenation-illegal voters than she does of the miscegenation-legal voters, and she is the only candidate for whom this is true,.
April 8th, 2011 at 1:46 pm
Numbers 12:1 – 12:13
April 8th, 2011 at 8:29 pm
Wow – those are some small Ns. I wonder if any number of the “liberal” republicans are black?
April 9th, 2011 at 12:41 pm
I wish they’d defined what a ‘liberal Republican’ is in this context.
Somebody who thinks the death penalty isn’t always merited?
April 11th, 2011 at 6:06 am
[...] points to the fact that racism is alive and well in the “postracial” United States. Gene Expression | “I think the liberal Republican opposition to interracial marriage in Mississippi is due [...]
April 11th, 2011 at 9:22 pm
You people do realize that they only polled 400 people, right? Hardly an indicator of what an entire state feels.
April 11th, 2011 at 10:34 pm
sample size objections are weak and generic. pollsters try and weight demographically. also, it’s not “what an entire state feels.” it’s a very narrow slice, WHICH YOU WOULD BE AWARE OF IF YOU BOTHERED TO READ THE POST AND THE COMMENTS.
April 12th, 2011 at 9:30 am
While Mississippi’s comment is off base, it is appropriate to recall that the margin of error for small subsamples, like the 16 person subsamples for “liberal” and “somewhat liberal” Republicans who are usual primary voters in Mississippi is very high (if the result were 50-50, which isn’t far from the mark for the sample as a whole or the “liberal” category, the MOE would be +/- 24.5% for a sample of sixteen). The difference in the percentage between the “liberal” and “somewhat liberal” categories, and the other categories is probably not statistically significant at the 95% level, as a result of the small subsample sizes.
The margin of error for the entire sample of 400 is +/- 4.9%.
It is also appropriate to wonder if “very liberal” respondents to this particular poll aren’t simply lying to pollsters in an attempt to manipulate the result.
April 12th, 2011 at 12:59 pm
[...] Razib Khan at Discover Magazine Blogs has taken a closer look at the numbers and challenged them “I think the liberal Republican opposition to interracial marriage in Mississippi is due to small sample size and mistaken responses”. [...]