The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
The Guiding Principles and New Rules document is now in effect.
The General Social Survey and Racism
Brian888
Registered User regular
Registered User regular
A friend of a friend on Facebook keeps trying to flog me with the following article:
http://dailycaller.com/2011/03/15/ron-schiller-is-wrong-people-who-want-small-government-are-not-racist/2/
The gist of the article is that according to the GSS, Tea Partiers and small-government conservatives are statistically less racist than liberals, especially white liberals. I suspect that there are flaws in this article, perhaps in the analysis of the statistical evidence, but I'm not enough of a statistician to analyze this myself. Is this article on the up-and-up, or are there flaws that I'm missing?
http://dailycaller.com/2011/03/15/ron-schiller-is-wrong-people-who-want-small-government-are-not-racist/2/
The gist of the article is that according to the GSS, Tea Partiers and small-government conservatives are statistically less racist than liberals, especially white liberals. I suspect that there are flaws in this article, perhaps in the analysis of the statistical evidence, but I'm not enough of a statistician to analyze this myself. Is this article on the up-and-up, or are there flaws that I'm missing?
Brian888 on
0
Posts
The Daily What?
Well let me go look at some of their other articles
...
...
...
Oh yeah these guys are clearly unbiased :rotate:
Hey everybody let's redefine torture to: "PROTECTING OUR CITIZENS"
This is an opinion they are seriously posting on their site
Give me a fucking break
When a non-partisan-rag source uses their so-called study then maybe we can talk about this
EDIT:
Hahahaha
We're not racists! We're totally 100% PC! We say African-American and not that other word! Totally not racist!
Rigorous Scholarship
That's what I'm trying to get at. Does the study say what he says it says, or not?
I feel the same way whenever talking heads bombard me with talking points. I feel like I'm much more well-informed than the average person, but of course I don't have all the latest statistics memorized, so I can never be completely sure what's a lie and what isn't. It's a very frustrating feeling, when you're 99% sure that someone is lying, but you can't be completely sure without doing way more research than it's worth.
What gives me pause is that the website didn't conduct the research. It published an article by a guy interpreting independent research.
Blue is white liberals, white is white conservatives, blacks are black, Latinos are brown, Asians are yellow, Natives are red... this graph is kind of racist itself!
If that's the case, this just means that White Conservatives are less likely to be both racist AND stupid - not a helpful metric.
RACDIF2: Categorical (Single)
(On the average (Blacks/African-Americans) have worse jobs, income, and
housing than White people.)
Do you think these differences are ...
Because most (Blacks/African-Americans) have less in-born ability to learn?
Categories:
{yes} Yes
{no} No
{dontknow} DON'T KNOW
{refused} REFUSED
From here.
It doesn't mean that there isn't some creative parsing of the data going on, but even if they're just using the yes/no responses it's a surprising result. Good thing though is that all the graphs appear to be steadily declining.
edit: ah yes, I was wondering about the latino and asian views. Still surprised by the 'black' result, though. That's awful.
The question this author is talking about reads:
You're allowed to answer more than one. The latest numbers on the NORC website for specific questions searchable by subject are from 2006. 40.1% of Americans surveyed answered YES on A. 14.3% answered YES on B. 49.1% answered YES on C.
I don't have the time or the software at home to run a correlation analysis for that many variables (white, Republican, small government, Blacks lower SES), but I'm skeptical based on the fact that his "others" number is lower than the total number. I also, frankly, am wondering what small-government Republicans did answer for this question (well, no, not really: it must have been "none of these" or just "no" on all 3) or what their answers would be. There are other questions indicative of racism that this author didn't explore, like whether respondents avoid driving through black neighborhoods (38.8% yes), or whether respondents believe that black people are lazy (35.8% in the top 3 boxes compared to 10.9% for whites in the top 3 boxes), or whether respondents anticipate special problems in workplaces with white workers and black supervisors (49.4% say yes). There are lots more.
TL;DR i'm not sure about this guy's methodology but that's definitely not the only GSS question he could have/should have used when making the claims he's making.
Rigorous Scholarship
Yeah, that's not good. It's hard to learn if you're convinced you can't learn.
Rigorous Scholarship
Fig.1 Years of Education Vs Goverment Action.
Since the GGS takes from 18 plus so a direct comparison is bad. 13.9 and 13.3 are really close and without variance we don't know if they are statisitically different. The standard way to deal with education is binning things rather than a continous scale, since you often find the high and the low end of the education scale being pro government whilst anti government is in the middle.
Figure 2 Governent does to much vs Blacks can't learn. <15% of any survey of the general public is getting into the lying/saying what they think they should say/joke/didn't get the question range/general insanity, and considering how much it bouces around earlier the temporary divergence at the end is suspect. Plus 12% could very easily be explained by old people rather than the general trend, and a proper analysis would have controlled for age groups IMO.
Figures 3 and 4: See the problems with Figure 2. Again with not showing the damn variance. I warrent this is just old people dying off in both parties, and seems to overlap one hell of a lot.
Overall seem pretty statistically trashy.
Edit: hah, beating by a million other posts.
A good metric I use here is to see if it's reported across multiple sources or not. If it's only reported by one hugely biased source, it's probably total bullshit. If more than 1 "reliable" news source posts it, it may be more worth your time to look deeper into.
Because yeah, I also don't have the time in my day to fact-check every absurd claim reported on every fucking batshit insane tabloid right or left-wing rag. There are crazies on both sides of the aisle, and even if there appears to be a study that would bolster my own position on something, I like to check to see if it's being picked up by a lot of outlets.
I'm a Canuckistanian socialist working on my second Master's degree, and I vote NDP, which makes me a highly educated left-winger even for a Canadian, but I would probably answer a 4 on that question. I'm not a big promoter of small government, I'm sure as shit not a Teabagger, I support welfare programs and subsidized education and all that standard hippie bullshit... but I do think that the American government does too much in some areas. I think Homeland Security is invasive security theater that provides the illusion of safety at the expense of liberty, and violates individual rights. On a similar note, I think forcing all airports to use the TSA instead of private companies for security is an unnecessary infringement on the free market, especially given the TSA's abysmal track record and ever-increasing degree of examination. I think governmental legislative attempts to strongarm unions (like for instance in Wisconsin) are incredibly inappropriate. I think the nation-wide standardized testing mandated by No Child Left Behind has significantly harmed the overall state of education in America. I think companies that drive themselves into bankruptcy through greed and bad business practices don't deserve to be bailed out with tax dollars.
I'm so far left on most issues that I'm a goddamned Communist compared to most Americans, yet there are numerous areas in which I do indeed think that the government is interfering too much. But that doesn't make me a Tea Party member, and trying to shoehorn me into that box based on my response to a single vaguely-worded question is just terrible fucking science that completely invalidates whatever "results" the authors claims to have found.
Seriously, that is just awful, awful research.
Yeah, but you really went more in-depth about how he statistically fucked it up. Nice work.
this seems reasonable.
It's because the white category is split into liberal white and conservative white. If you combine the two, the number of whites who believe blacks have less in-born ability to learn is probably higher.
It has about the same amount of meaning as saying nonconservatives like wasteful spending. By that metric, everyone is conservative, because no one supports spending they think is wasteful.
You can download the data yourself at http://www.norc.uchicago.edu/GSS+Website/Data+Analysis/
I'm doing some regressions. I'll get back to you in a little while.
Or you can export the data and do it yourself.
Edit: never mind, I was waaaayyyyy beaten.
the "no true scotch man" fallacy.
Honestly, without even looking at the article, I'd place 10 to 1 odds that it's the usual bullshit "Liberals perpetuate the welfare system which disproportionately affects blacks and keeps them from hauling themselves up by their bootstraps. Ergo: Racists!"
Oh look at that. Conservatives don't think blacks are inherently inferior, they're just lazy!