Gene Expression

Which demographics know science?

The always fascinating Inductivist takes a look at science comprehension of Americans via the GSS. Here’s his methodology:

In 2006 the General Social Survey asked 437 respondents eleven basic science questions. The first one, for example, was whether the earth’s center is hot. I gave each person one point for answering a question correctly, and then summed the scores. My next step was to convert these totals so they resemble IQ scores. I set the white mean at 100, and the standard deviation at 15. Here are some averages:

Since we know the standard deviation I decided that it might be interesting to take the numbers he generated and convert them into standard deviation units, and display them on one single chart in a comparative manner.

i-3eed7ee01e73822c06b0d91e12dfced8-sciencedemo.jpg

The first thing you might ask is how can both male and female “science IQ” be below average? Remember that the scale is standardized to the white average, and as you can see both blacks and Hispanics have lower scores than whites, resulting in an total population average score below that of the white median. So with that out of the way, what do we notice? Nothing much that should surprise us. Blacks and to a lesser extent Hispanics know less about science than whites. But since science is logocentric and linear that makes sense, proficiency is directly proportional to white cultural content. So Hispanics, who are generally of part European cultural heritage do better than blacks. Men do somewhat better than women, probably because of the patriarchal structure of modern science, which discourages female involvement. Knowledge of science is directly proportional to education; ‘nuf said. Since around 1/3 of American Nobel Prize winners in science are Jewish one would expect that Jews, as a whole, have a high science IQ. But what’s with Protestants doing so much worse than Catholics? Part of it is plain demographics; blacks are more likely to be Protestant than Catholic and do worse on this test than whites. On the other hand, Latinos are more likely to be Catholic than Protestant, and also do worse, though not as badly. Nevertheless, I think ‘tarded Protestant Fundamentalists are making their cretinism known in these statistics. I suspect that that’s why those who attend church more than once a week are so science ‘tarded; such hyperattendance is a feature of Fundamentalists.

To me the regional data is the most fascinating. Here is the rank order from the most science ignorant to the least:

South Atlantic
West South Central
East South Central
East North Central
Middle Atlantic
Pacific
New England
Mountain
West North Central

It seems to me that the states which are the most savvy about science are those who tend to be part of what Michael Lind has termed Greater New England:

If you look at a linguistic atlas of the United States, you’ll notice something striking. The “Upper North” dialect zone identified by students of American speech patterns is almost identical to the blue-state zone on the Electoral College map: New England, the Great Lakes states, and the Pacific Northwest. This is “Greater New England” — the regions settled by New Englanders and their descendants from the 17th to the 19th centuries.

This is the region where Barack Obama has done particularly well:
i-1189a2fc5126b57d7019154064af6e42-baracknewengland.jpg

i-21058e6c48e403d743614ff966ed72f1-evolvsscienceIQ.jpgSince this is a biology focused weblog, I decided to take the regional science knowledge data and just plot them vs. the proportion who accept human evolution in any region. I’m assuming these data won’t shock anyone’s priors. Southerners are the least intelligent regional subculture in these United States, and, they reject descent from apes and exhibit a more broad ignorance of scientific questions.

Just out of curiosity I looked up a few other of the above demographics re: belief in human evolution (e.g., male vs. female) and added them to the regional plot, and this is what I got:
i-d72b9dc014b77fc6387b2c96f025d56a-sciencedemoX.jpg

I’ll leave it to readers to figure out what the labels for each dot are (took me no more than 5 minutes to find the polls).

Comments

  1. #1 John Emerson
    July 9, 2008

    See, everyone really is above average in Lake Wobegon.

  2. #2 D.V.N
    July 9, 2008

    While I have recently begun to read your blog regularly (really excellent), I was struck by a couple of your formulations in this in general very interesting installment.

    “Blacks and to a lesser extent Hispanics know less about science than whites. But since science is logocentric and linear that makes sense, proficiency is directly proportional to white cultural content”

    That sounds like a lot of postmodernist woo simply taken for granted, doesn’t it? What about adjusting for the various parameters associated with educational level (and parents’ educational level), for instance (which, as far as I understand, the original survey didn’t). I’ve read the equivalent of that way of putting it before, with the writers sincerely expressing their opinions, and unless it is combined with a radical relativism about truth (e.g. Rorty-style), it is more or less tantamount to saying, in a vague and politically correct-sounding way, that blacks are less intelligent than whites.

    Or am I being Poed here (or simply misunderstanding you) insofar as you do highlight the educational factor later on, following a hilariously simplified (and – possibly – aptly sarcastic) reduction of the importance of gender roles and societal expectiations in gender differences in science? “Men do somewhat better than women, probably because of the patriarchal structure of modern science, which discourages female involvement. Knowledge of science is directly proportional to education; ‘nuf said.” There are different ways to read your point here, I admit.

    Anyway: fascinating post.

  3. #3 razib
    July 9, 2008

    that blacks are less intelligent than whites.

    i would contend that “Intelligence” is a white male construct. other groups must be judged by their Own Ways of Knowing.

  4. #4 D.V.N
    July 9, 2008

    “i would contend that “Intelligence” is a white male construct. other groups must be judged by their Own Ways of Knowing”

    Well sure, but I was intending “intelligence” in very loose sense, more like “ability to represent and understand one’s environment and reacting to it in an efficient manner”. “Other ways of knowing” is subsumed under this ‘definition’, since knowing that p (as opposed to imagining or merely believing that p) requires, at the very least, the truth of p (so this “other way of knowing”, whatever it is, must be some way of disclosing truths)

    The point is really the following argument:

    Premise 1: Blacks and to a lesser extent Hispanics know less about science than whites, and this is because science is logocentric and linear, and hence proficiency is directly proportional to white cultural content.

    Premise 2: Science discloses the ultimate structures of reality, and its success at arriving at powerful models of reality (as opposed to any other ways, e.g. the study of religious scriptures), shows that the methods of science, the “logocentric and linear” thinking, give us the best available tools for disclosing truths.

    Conclusion: Blacks and to a lesser extent Hispanics are less proficient at disclosing and understanding the real structures of reality.

    Now, if you don’t like that conclusion, but want to commit yourself to premise 1, it seems like you are very much committed to rejecting premise 2. For my part, I endorse the second premise, but reject the conclusion since I have serious doubts that the claim made by the first premise is even intelligible (far less true) – “other ways of knowing” would have to be judged according to the common, reasonable standards for justification and confirmation (at the very least be information gained by a method which reliably arrives at truth), else I have no idea what it is supposed to mean. Claiming that people from other cultures generally imagine and intuit a lot of stuff, while it is the province of white males gain knowledge of the world, hardly sounds any better to me than saying that they are less intelligent.

    A remaining possibility would be that the subject matters of these “ways of knowledge” are different (for instance, ethics and aesthetics aren’t generally the subject matter of science), but this looks like a long shot.

  5. #5 razib
    July 9, 2008

    Premise 1: Blacks and to a lesser extent Hispanics know less about science than whites, and this is because science is logocentric and linear, and hence proficiency is directly proportional to white cultural content.

    interpolation: and, logocentric and linear thinking is characteristically the product of white male culture.

    Science discloses the ultimate structures of reality, and its success at arriving at powerful models of reality (as opposed to any other ways, e.g. the study of religious scriptures), shows that the methods of science, the “logocentric and linear” thinking, give us the best available tools for disclosing truths.

    what “reality” and “truths” is this you speak of? if reality and truths lead to the construction of Material Objects via Technology, which itself is embedded in a Free Market, then yes. but i would contend that the whole set of Reality and Truth is intersected* only minimally by reality and truth as understood by the white male paradigm. conflating the subset for the whole set is fundamentally Hegemonic and dismissive of Other Ways of Knowing.

    * i apologize to those outside of the white male patriarchy who find my set theoretic allusions obfuscatory!

  6. #6 Matt McIntosh
    July 9, 2008

    Chet Snicker has stolen Razib’s identity!

  7. #7 chet snicker
    July 9, 2008

    Chet Snicker has stolen Razib’s identity!

    sir,

    i object to the implication of your assertion in regards to my character! i am no subaltern who can not speak, who lacks Voice.

    yours truly,
    c.v. snicker

  8. #8 blah
    July 9, 2008

    “i would contend that “Intelligence” is a white male construct. other groups must be judged by their Own Ways of Knowing” Well sure,

    BWAHAHAHAHAHAAAAAAAAAAAAAAAAAAAAAA…

    (weeps laughing)

    “Well, sure”

    I can’t get enough of that. Now we know how Sokal got his article through.

  9. #9 Onkel Bob
    July 9, 2008

    Wait a second. If these are indeed the questions then the poll does not test scientific literacy but determine whether the person is sentient.
    Mind you, some of the questions are poorly worded to the point of vagueness.

    The universe began with a huge explosion. (Is that true or false?)

    Does this mean the universe did not exist before the big bang?

    Electrons are smaller than atoms. (Is that true or false?)

    Given that electrons are subatomic particles, is not the answer obvious? However, you’re leading the respondent to the idea that electrons are not part of the atom.

    That said this whole thing smells of Poe to me. Those questions should be on “Are You Smarter Than a 5th Grader?”
    However, considering who has been elected to the office of president for the last century, it’s obvious that the electorate is not smarter than a 5th grader…

  10. #10 gss
    July 9, 2008

    data is here: http://sda.berkeley.edu/cgi-bin32/hsda?harcsda+gss06

    key variable is scitotal

    for posterity, scitotal = hotcore2 + radio2 + boygirl2 + lasers2 + electro2 + virus2 + bigbang2 + condrif2 + evolved2 + earthsu2 + earth2

    race variables is raceth

    the minimum score was 4 correct. the percentages for 4-11 correct for each group as follows:

    white non hispanic
    4: 0.00%
    5: 2.54%
    6: 3.66%
    7: 9.30%
    8: 16.34%
    9: 20.85%
    10: 17.46%
    11: 29.86%

    black non hispanic
    6.38%
    19.15%
    21.28%
    27.66%
    12.77%
    4.26%
    4.26%
    0.00%

    hispanic
    0.00%
    0.00%
    15.79%
    31.58%
    15.79%
    5.26%
    26.32%
    5.26%

    zero individuals in the black group got all 11 correct. the hispanic group is clearly bimodal.

    in a multiple regression, raceth and degree are both significant.

  11. #11 razib
    July 9, 2008

    in a multiple regression, raceth and degree are both significant.

    the validity of multiple regression presupposes a LINEARITY which is not a position that all would admit. rather, this short of Analysis is a classic example of White Males attempting to play out their Hegemonic cognitive narrative. one can not tear down the Master’s House with the Master’s Tools!

  12. #12 gss
    July 9, 2008

    the “black non hispanic” + graduate degrees score on average (7.25 correct) worse than the “white non hispanic” + less than high school (7.43 correct)

  13. #13 D.V.N
    July 9, 2008

    “what “reality” and “truths” is this you speak of? if reality and truths lead to the construction of Material Objects via Technology, which itself is embedded in a Free Market, then yes. but i would contend that the whole set of Reality and Truth is intersected* only minimally by reality and truth as understood by the white male paradigm. conflating the subset for the whole set is fundamentally Hegemonic and dismissive of Other Ways of Knowing.”

    What the … Now I AM being Poed, am I?

    I wasn’t speaking about any particular “reality” or “truth”. I was merely claiming that whatever these are, science provides us with the best way of accessing them. You haven’t addressed this point except for invoking the very strange and unexplained “other ways of knowing”. Even white males tried other ways earlier on, most spectacularly, perhaps, in the case of the medieval neo-platonists. They didn’t KNOW different things than we do. They knew LESS than we do, but had a lot of false believes about reality, however that notion is spelled out in the end. They were WRONG about a lot of stuff about which we later obtained a better understanding. We have improved our methods for arriving at truth. That was my claim. Apparently – and unless I am getting POED – you disagree (although your talk about white males having access to only a subset of the set of truths makes it a little fuzzy: “the earth moves” would be a member of such a set while “the earth doesn’t move” would not be a member – thus suggesting that the subset entertained by present-day white males IS more extensive than the one entertained by white males 700 years ago, and hence that some methods for compiling such a set are better than others – if you deny that we are currently better at arriving at truths, then you are indeed a radical relativist about that notion (which subsequently makes the claim that other groupings have different methods for arriving at (different?) truths incoherent)). I still haven’t committed myself to any particular notion of truth or reality. But you claim that other groups have “other ways of knowing”. Can you explain what truths (rather than falsehoods) they arrive at through these “ways” – and account for how these “ways” are supposed to disclose truth (in whatever sense you like) rather than falsehoods (as medieval students of the scripture so fabulously failed to)? That is, how does this “other ways of knowing” distinguish themselves from mere imaginings?

  14. #14 razib
    July 9, 2008

    That is, how does this “other ways of knowing” distinguish themselves from mere imaginings?

    have you ever been in a drum circle? that is a Way of Knowing, a Way of Living, of Acting and not Thinking on the world. it is a way of accessing the Truth of the Ground of Being.

  15. #15 agnostic
    July 10, 2008

    hotcore2 + radio2 + boygirl2 + lasers2 + electro2 + virus2 + bigbang2 + condrif2 + evolved2 + earthsu2 + earth2

    OMG, those are some of the dirtiest variables ever!

  16. #16 D.V.N
    July 10, 2008

    “have you ever been in a drum circle? that is a Way of Knowing, a Way of Living, of Acting and not Thinking on the world. it is a way of accessing the Truth of the Ground of Being.”

    This suggestion involves a category mistake. There is no object, “the Truth” out there which we access by different means. Truth is a property of claims/propositions/statements (or the like), a property instantiated by the claim “the earth is appr. 4.5 billion years old” and not instantitated by the claim “the earth is not 4.5 billion years old”. Truth isn’t an object or some quasi-religious entity (lest you owe everyone an explanation of this rather wild metaphysical claim – viewing truth as an object is not totally without precedents; Frege viewed truth as an abstract object, a value obtain by the saturation of an object (or sense) to a function-expression (a concept), but this can hardly be what you are aiming at).

    Look, I’ve done my share of meth and gasoline huffing. It was fun and certainly altered my mental states/cognitive faculties for a finite period of time. I would not dream of claiming that it provided me of some different means of gaining knowledge, however (with knowledge understood as true information obtained by a method which reliably distinguish true claims from false ones). Drum circles don’t impress me any more than those prayer circles of religious fanatics (from which some participant also claim to have gained certain “insights” – they haven’t, of course. Knocking myself in the head with a hammer could theoretically cause me to entertain some new ideas as well, like believing myself to be Napoleon. It is not a way of gaining knowledge. Delusions and knowledge are polar opposites. “Method that reliably distinguishing true claims from false ones” is the crucial point; this is what makes a belief justified, what gives you a REASON for believing something (which is required for anything to count as knowledge). Mystical experiences don’t give you knowledge unless you can identify some justificatory structure apt for sorting the true from the false claims, and that structure just isn’t part of the experience itself.

    Where does the drum circle fit here? Using your set metaphor, what members of the complete set of truths are obtained by engaging in these activities, which are not in the “subset” obtained by scientific inquiry? And what makes them count as knowledge?

    I don’t really want to be this dismissive (cause I do like this blog), but the whole thing about “other ways of knowing” sounds like PM woo.

    Why does PM notion still seem attractive to some? The maasai believed humanity to be fashioned by the Creator deity from a single tree or leg which split into three pieces. Dead white male scientists found that humans have evolved through natural selection. Dead white male scientists were right, the maasai were wrong. Score: Dead white male scientists 1, maasai 0. This result apparently seemed unacceptable to some, especially politically informed European apologists for the atrocities committed by Europeans towards other cultures, so instead of this way of viewing it, they entered the stage with the feeble “look, the maasai are equally correct, they just live under a different conceptual scheme/different ways of knowing/access to some different part of reality”. Quite a feeble and cowardly way to try to explain away disagreement and the fact that one party appears to be right and the other wrong, isn’t it?

    The next step in PM ensued from recognizing that not everything goes – we don’t want to be nihilists about truth. Rather, let us say that different conceptual schemes have different standards for truth. By itself, this wouldn’t get the fact that the dead white males were right and the maasai were wrong out of the way. In order to explain away the disagreement, enter incommensurability of different conceptual schemes – claims made under one conceptual scheme cannot be evaluated by standards of truth in another.

    But of course – the claim that there are incommensurable conceptual schemes itself … what standards for truth is that one supposed to be evaluated against? It would have to be a transcendental standard, exactly what proponents of truth-relativism are committed to denying the existence of. If it should be evaluated against no standards, we would have absolutely no means for determining what on earth the claim amounts to. Hence, truth-relativism is an incoherent position, and so is the notion of incommensurable conceptual schemes.

    In the end, then: Could anyone give me a member of the set of truths which is not a member of the dead white male accessible subset? Or a reason to think there are genuine such truths?

  17. #17 Beowulff
    July 10, 2008

    I wonder if the statistics wouldn’t be more meaningful, more informational and more akin to IQ scores if they had been normalized to the whole population, instead of to one group? Especially if the group in question appears to be an outlier.

    For instance, looking at the bar graph, my expectation was that it would be reasonably balanced around the average, with about as much bars going to the right as going to the left. However, most bars by far go to the left, which puzzled me at first. You can only understand this after reading the non-standard and non-intuitive way the scores are defined.

  18. #18 razib
    July 10, 2008

    nothing’s stoppin’ you beowulff! ;-) in any case, whites are the majority of the sample. so i think it’s a defensible standard.

  19. #19 deadpost
    July 10, 2008

    Hence, truth-relativism is an incoherent position, and so is the notion of incommensurable conceptual schemes.

    Do people actually believe in truth relatism? Or are they acting it for political correctness… ie. saying, its ok, “we know you are wrong but we’ll act like you’re right anyways to make you feel better”… that’s treating like children.

    By the way, there is no way to argue around the self-contradiction of truth relativism (ie. If there is no one truth, how do we know that truth is true), is there? No matter how those dumb (“fake”) philosophy majors claim.

  20. #20 deadpost
    July 10, 2008

    Where are all the South Asians?
    Whoah… they must be so many standard deviations right that they couldn’t fit on the graph ;)

  21. #21 sinmantyx
    July 14, 2008

    Am I missing something? Humans did not descend from apes. The idea that “animals” and “humans” are mutually exclusive also makes no sense. Are those your errors, or errors in the questions themselves?

    Perhaps that could by why the points in that particular graph could be modeled as some sort of parabola-like curve. If you know TOO much, you will (ironically) answer the same way as someone who knows very little or rejects science all together.

    The conjecture that women tend to do worse than men only because they are oppressed and minorities do poorly because of cultural context a little bit puzzling. The fact that women differ from men cognitively is pretty well established. You aren’t going to hurt my feelings by pointing that out.

    I find race-related statistics generally unnecessary and provocative, partly because doing research concerning a trait that is essentially a perception – that can have little to do with a person’s culture, education, background, or even the VAST majority of their DNA – just doesn’t seem reasonable. Terms like “Hispanic” and “African-American” don’t even refer to race, but are generally used as if they do.

    Maybe that’s why I went into physics and not sociology. I couldn’t stand how messy it all becomes. Then you also have the issues of trying to find causal relationships when all you have is a nest of confound-ridden correlations.

    Also, where’s the Asians? I’m not the first one to ask. Perhaps they were all Hispanic and checked that box? :)

    Another reason I try not to put too much stock in these types of studies is that it is very difficult to get sample spaces large enough to really do the job. There may have been 437 respondents, but once you break that larger group into smaller groups – you inevitably end up with an even smaller number attempting to represent a VERY large and diverse group.

    In the end, I don’t see the point. Each student that comes into my classroom and each person who asks me a question about science, is a sample space of one. What’s the point of making gross generalizations? I don’t understand the appeal.

  22. #22 ray
    July 15, 2008

    sinmantyx,
    Am I missing something? If humans and chimps are late branches in the superfamily ‘apes’, what would you call the last common ancestor of the two?
    I agree with your complaints about fractioning a moderately sized sample, and the racial distinction; the regional and religious comparisons are interesting and more reliable.

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.