Interesting interview with Lance deHaven-Smith, a professor specializing in Florida election law and the author of a new book about the 2000 Florida election. From the interview:
LdHS: It’s an embarrassing outcome for George Bush because it showed that Gore had gotten more votes. Everybody had thought that the chads were where all the bad ballots were, but it turned out that the ones that were the most decisive were write-in ballots where people would check Gore and write Gore in, and the machine kicked those out. There were 175,000 votes overall that were so-called “spoiled ballots.” About two-thirds of the spoiled ballots were over-votes; many or most of them would have been write-in over-votes, where people had punched and written in a candidate’s name. And nobody looked at this, not even the Florida Supreme Court in the last decision it made requiring a statewide recount. Nobody had thought about it except Judge Terry Lewis, who was overseeing the statewide recount when it was halted by the U.S. Supreme Court. The write-in over-votes have really not gotten much attention. Those votes are not ambiguous. When you see Gore picked and then Gore written in, there’s not a question in your mind who this person was voting for. When you go through those, they’re unambiguous: Bush got some of those votes, but they were overwhelmingly for Gore. For example, in an analysis of the 2.7 million votes that had been cast in Florida’s eight largest counties, The Washington Post found that Gore’s name was punched on 46,000 of the over-vote ballots it, while Bush’s name was marked on only 17,000.
RinR: For your research, you merged this set of data with detailed profiles of Florida’s electoral precincts. What did you find?
LdHS: One of the things I found that hadn’t been reported anywhere is, if you look at where those votes occurred, they were in predominantly black precincts. And (when you look at) the history of black voting in Florida, these are people that have been disenfranchised, intimidated. In the history of the early 20th century, black votes would be thrown out on technicalities, like they would use an X instead of a check mark.
So you can understand why African Americans would be so careful, checking off Gore’s name on the list of candidates and also writing Gore’s name in the space for write-in votes. But because of the way the vote-counting machines work, this had the opposite effect: the machines threw out their ballots.
I know that a lot of people are sick of hearing about ballot-counting in Florida in 2000. I’m not. To a great extent, my belief that the US has a working democracy was shattered in that election. So, also, was my belief that high-level Republicans ever act in good faith (before the 2000 election, I actually had some admiration for Scalia).
Not that I really beleive that Democrats as a whole would have acted better. (In the interview, deHaven-Smith argues that the specific Democrats who were an alternative to Jeb Bush in Florida in 2000 would have acted better, had they been in charge of vote counting; that may be the case, but I don’t believe we can generalize from those specific Democrats to Democrats in general).
DeHaven-Smith argues, persuasively, that the real problem in Florida wasn’t just bad technology; it was a system in which partisans with a strong stake in the outcome of elections, are in charge of administrating elections, and also in charge of investigating problems afterwards. This creates a strong bias against both fair elections, and very little motive for anyone to strive for absolute honesty in vote-counting.
Curtsy: Kevin Drum.
Pingback: Just Between Strangers
Pingback: feminist blogs
Part of the reason why so few people have brought up the issue of race and overvotes is that it would reveal that blacks were more likely to spoil their ballots than whites. Despite claims that the voting systems used more frequently in black districts (e.g. punchcards vs. optical scanners) were more prone to error, the fact of the matter is that even within a county [within which a single method would be used] there were more spoiled ballots in black precincts than white precincts.
That is, even using the same system, black voters were more likely to spoil their ballots than whites.
Talking about the higher rate of black spoiled ballots would therefore bring up all sorts of nasty Bell Curve-type issues, which caused most people to avoid it like the plague.
La Griffe du Lion has an analysis of the issue.
Talking about the higher rate of black spoiled ballots would therefore bring up all sorts of nasty Bell Curve-type issues, which caused most people to avoid it like the plague.
La Griffe du Lion has an analysis of the issue.
It’s interesting how people who are determined to believe that melanin can be used as a predictor of intelligence will come up with reasons for believing it, even in response to a post which clearly gives a far more rational reason for “spoiled votes” than La Griffe du Lion’s notion that the more melanin you have, the stupider you are. See also The Mismeasure of Man, Stephen Jay Gould, for a fascinating analysis of the historic use of science to “prove” that whites are smarter than blacks, men are smarter than women, richer people are smarter than poorer people, etc. (Go for the second edition: that’s the one with the chapter explaining how The Bell Curve is bad science.)
even in response to a post which clearly gives a far more rational reason for “spoiled votes” than La Griffe du Lion’s notion that the more melanin you have, the stupider you are.
I don’t think that this post gives a “far more rational reason” why black voters were more likely to spoil their ballots than white voters (i.e., black voters were being extra careful). As a reason, this is mostly speculation.
In fact, the major point of Amp’s post isn’t about why black voters were more likely to spoil their ballots, the major point is that the system was unfair because it disqualifies spoiled ballots even when the intent of the voter was unambiguous; and that the system of reviewing the elections was unfair because
the people involved had no motivation to question the system’s pedantic insistence that ballots be invalidated even when there is clear intent.
What would be your guess as to how Gore would have handled 9/11 and Iraq?
Glaivester: I don’t think that this post gives a “far more rational reason” why black voters were more likely to spoil their ballots than white voters (i.e., black voters were being extra careful). As a reason, this is mostly speculation.
Of course it is. But it’s speculation based on solid facts: whereas the speculation you cited, that the more melanin you have, the stupider you are, is not based on facts at all: it’s based on bigotry. (It’s not difficult, I admit, to find a “far more rational reason” that the unscientific nonsense of The Bell Curve, so this is not much of a compliment.)
In fact, the major point of Amp’s post isn’t about why black voters were more likely to spoil their ballots
Yet it was the only point you focussed on in your first comment on it.
But it’s speculation based on solid facts: whereas the speculation you cited, that the more melanin you have, the stupider you are, is not based on facts at all: it’s based on bigotry.
For one thing, it is based on the fact that people of African descent score lower on IQ tests on average than people of European descent. So it is based on facts, whether or not you agree with the way those facts were interpreted.
Also, again let me say that race is not and has never been about skin color. It has to do with ancestry. People use “whites” as short-hand for people of European descent, and “blacks” as shorthand for people of African descent, because skin color is the most obvious difference, but skin color is not itself what people are referring to when talking about race.
(For example, despite people referring to people of East Asin origin as “having yelow skin,” the main feature by which people distinguish Europeans from East Asians is actually eye shape, not skin color).
In any case, back to my original point:
Lance deHaven-Smith is obviously distressed by the fact that the mainstream media and the people in power have ignored the issue of the overvotes, and by the fact that they have also ignored that the overvotes were mostly in predominantly black precincts.
In terms of this issue (i.e., why the gap wasn’t publicized), what is important is not whether there is a black-white cognitive gap or not. It is whether or not people think that there is one, and how this report would affect people’s thinking on the issue.
Why the Republicans and conservatives were not interested in this fact is obvious. They don’t want to publicize anything that suggests that they didn’t win fair and square. The question is, why didn’t Democrats and liberals, or non-partisan groups, publicize it more?
The fact of the matter is, that there was no easy way to bring up this issue without bringing up the fact that voters of African descent were disproportionately likely to spoil their ballots, even using the same voting methods. And whatever the actual reason for this, it is difficult to bring up the fact that African-Americans were disproportionately likely to spoil their ballots without on some level it sounding like you are saying that black people can’t figure out how voting works. That is, whether or not IQ had anything to do with the discrepancy, there is no way to bring up the discrepancy without having to hash out the issue of race and IQ, and without, in the shot term t least, giving ammunition to those who believe that there is a cognitive gap.
So why didn’t more people on the left bring this up and then launch a rebuttal of the idea that the higher percentage of spoiled ballots were due to any cognitive differences?
Here are three possible reasons:
(1) Because they felt that it would be easier just to ignore the issue.
(2) Because they were afraid that regardless of how much evidence they could muster to show that innate cognitive differences were not the cause of the discrepancy, all that the majority of people would remember is that “black people hav a harder time filling out a ballot correctly.”
(3) Because they themselves actually believe that black people aren’t as capable, and don’t have faith that they would be able to refute those who claim cognitive differences.
It is important to remember that whether or not there is a racial cognitive gap, there are probably a lot more people who believe that there is than are willing to admit it; and someone who believes that there is a cognitive gap but who feels that it is impolite to admit to it is going to be very unwilling to discuss anything that would touch on the issue.
Glaivester, that link you provided was not only written in a rather annoying and precious style, it was written by an absolute racist.
It’s true that tested I.Q.s are, on average, lower for blacks than whites. But to discuss this without mentioning the considerable evidence that I.Q. is neither innate nor an absolute measure of all cognative ability is to give comfort to the most deplorable racist beliefs in our society.
As for why blacks were more likely to overvote than whites, I’d say there are a lot of explanations – such as overcaution (as suggested in the quoted interview), educational level (which correlates with practice in filling in forms), white collar vs blue collar work (ditto), mentoring (did you have a parent or other experienced adult available to walk you through the process before the first time you voted?), etc., which should be considered before we leap to talking about IQ.
I’d go further than that. Florida law – “the system” – makes it clear that votes are to be interpreted whenever the voter’s intent is clear, which is certainly the case with most overvotes. But the system was administrated by people who were highly motivated to confound the rules, so that clear intent was ignored.
For one thing, it is based on the fact that people of African descent score lower on IQ tests on average than people of European descent. So it is based on facts, whether or not you agree with the way those facts were interpreted.
You really need to read The Mismeasure of Man. “IQ tests” are not neutral measurements of “intelligence”: they are culturally biased tests standardized towards the dominant group in that culture. That white Americans tend to score higher than black Americans (“people of African descent” vs “people of European descent” is inaccurate – a black American may have as much European ancestry as a white American) says only that IQ tests used in the US are standardized towards white American culture, not black American culture. There’s a good basic site here.
That is, whether or not IQ had anything to do with the discrepancy, there is no way to bring up the discrepancy without having to hash out the issue of race and IQ,
Only if you’re desperate to do so – if you so urgently want to believe that there’s a correlation between melanin and intelligence that you will pick on anything to prove your point.
More accurately, it might be said that given the number of racists in the US, and the respectability of racist ideas such as that melanin correlates negatively to intelligence (that is, the more melanin you have, the less intelligent you are, to summarize it), any time the subject of the number of spoiled ballots in black counties was brought up, the idea that this happened because black people are less intelligent than white people was bound to be raised. I know, because I saw it happen: every time the topic of the Florida election was discussed, except on Obsidian Wings, where the presence of a black Republican may have quelled such nasty talk before it could begin.
Florida has a long history of disenfranchising black voters. To any reasonable person, the obvious reason why far more ballot papers from black voters were declared “spoiled” than ballot papers from white voters, would be because, once again, Florida was carrying out a long-standing tradition of disenfranchising black voters.
It’s rather like – indeed, as Stephen Jay Gould points out, it’s exactly like – arguing that the reason women get paid on average less than men is because women are less intelligent than men – when there are a vast number of social reasons for the gender pay gap, all of which are inherently more probable than a correlation of male gender with intelligence.
When people leap to the conclusion that black voters are more likely to spoil ballots that white voters because black voters are more stupid, discarding unexamined all the more reasonable social reasons why more ballots cast by black voters might be declared spoiled – then I have to conclude that they do so because they prefer a racist conclusion to rational thinking.
I’d go further than that. Florida law – “the system” – makes it clear that votes are to be interpreted whenever the voter’s intent is clear, which is certainly the case with most overvotes. But the system was administrated by people who were highly motivated to confound the rules, so that clear intent was ignored.
True. I hadn’t read in the article when I made my comment.
More accurately, it might be said that given the number of racists in the US, and the respectability of racist ideas such as that [percentage of African heritage] correlates negatively to intelligence (that is, the more [ancestors of African descent] you have, the less intelligent you are, to summarize it), any time the subject of the number of spoiled ballots in black counties was brought up, the idea that this happened because black people are less intelligent than white people was bound to be raised.
And my point is that rather than trying to argue against that point, a lot of people are content not to bring it up in the first place.
Florida has a long history of disenfranchising black voters. To any reasonable person, the obvious reason why far more ballot papers from black voters were declared “spoiled” than ballot papers from white voters, would be because, once again, Florida was carrying out a long-standing tradition of disenfranchising black voters.
Only if it can be shown that ballots with the same mistakes were counted in white precincts and not ocunted in black precincts. Or, if it can be shown that blacks tend to make certain mistakes, and whites others, and the mistakes that blacks were more likely to make were more likely to be considered to spoil the ballot.
And my point is that rather than trying to argue against that point, a lot of people are content not to bring it up in the first place.
What I mean is that rather than arguing against the idea that there is a meaningful racial IQ gap, a lot of people just want to sweep the whole issue under the rug, and not to discuss any subjects (e.g., spoiled ballots in Florida) that might cause the issue to be brought up.
In any case, I would prefer a system with computerized voting, but that created hardcopy ballots that would be used in the counting. It would also be useful to have some way of linking the ballots to particular voters while maintaining their anonymity (e.g. having each voter get a voter ID number which was associated with their ballot and with their name, so that people could check up and make certain that each ballot was cast by a real registered voter [one could check the numbers on all of the ballots from the precinct and another could check all of the numbers against the names; no one would know all three and so no one could tell who voted for what]). This would make it harder to rig voting machines or otherwise add fraudulent votes.
The Mismeasure of Man is a text popular with lay persons and social scientists, however it is a text making claims of biology and psychology, and among the peer review of fellow biologists, psychologists, neurologists, and psychometricians, it has been criticized and refuted, as has Gould himself.
The late British evolutionary biologist John Maynard Smith, winner of the Crafoord Prize (issued by the same committee that selects Nobel Prize Winners) bemoaned about Gould,
Go visit this site.
Gould’s alleged debunking of methodologies like factor analysis is inaccurate, as those methodologies remain credible and valid in chemistry and engineering.
Go visit this site.
Harvard Medical School Professor Bernard Davis warned,
Go visit this site.
The Neuroscience division of the journal “Nature,” one of the most reputable scientific peer publications, has issued research demonstrating that general intelligence (g) is strongly correlated with brain structure and genetics.
Go visit this site.
Arthur Jensen, one of Gould’s prime targets in the book, responded to Gould’s misinterpretations and glaring errors.
Go visit this site.
If Jensen is such a demented racist, perhaps someone should notify the APA, which keeps publishing him up into 2005.
Go visit this site.
I wouldn’t call Jenson demented, but he – and his writing partner in the article you link to, Philip Rushton – are both hard-core racists who have dedicated their careers to denigrating Blacks.
In the scientific community, however, it’s the norm to allow debate; most scientists would not agree with the idea that because someone is a racist, he or she therefore should not be published. Therefore, your implied claim – that because Jenson had work published in a peer-reviewed APA journal, he’s not a racist – holds no merit. The APA has no rule against publishing racists. (Or against publishing the demented, for that matter, as long as their dementia doesn’t prevent them from writing a facially coherant article).
The particular publication you cite is one in which the editors were clearly trying to present a point/counterpoint forum; the issue included articles criticizing Jensen and Rushton, such as this one, this one and especially this one. The fact that the APA is providing a forum for debate is not evidence that Ruston and Jensen are not racists, nor is it an endorsement by the APA of their views. (One of their critics in the issue is a past president of the APA).
In addition to the above links, for more criticism of Rushton and Jensen, see this article by Tim Wise; this one from FAIR; and the online book Ressurecting Racism.
Dorset, this is such a confused criticism I rather think you don’t comprehend what you’re saying: Gould’s alleged debunking of methodologies like factor analysis is inaccurate, as those methodologies remain credible and valid in chemistry and engineering.
Gould made no attempt to “debunk factor analysis”. Rather, Gould was able, thanks to his knowledge of factor analysis, to show how factor analysis had been misused to try to “prove” that intelligence can be depictable as a single number and essentially immutable. You could not possibly read Gould’s chapter on The Bell Curve in The Mismeasure of Man and think that Gould was trying to debunk factor analysis, nor think that a statement such as “those methodologies remain credible and valid in chemistry and engineering” was anything but an absurd non sequitur. I guess, therefore, that your entire comment here is sourced not from anything you’ve directly read, either by Gould or by critics of Gould’s work who are responding to what Gould has actually written, but from some website which is incoherently posting anything that the website owner could find that looked like a criticism of Gould – including such an absurd assertion that Gould was trying to “debunk factor analysis”.
Glaivester Writes: And my point is that rather than trying to argue against that point, a lot of people are content not to bring it up in the first place. What I mean is that rather than arguing against the idea that there is a meaningful racial IQ gap, a lot of people just want to sweep the whole issue under the rug, and not to discuss any subjects (e.g., spoiled ballots in Florida) that might cause the issue to be brought up.
And some people, rather than arguing against the idea that there is a meaningful racial IQ gap, want to bring up the idea that there is such a gap, rather than discuss the subject of spoiled ballots in Florida and the institutionalised racism that has consistently, over many decades, in various different ways, sought to disenfranchise black voters in Florida and in other US states.
For example, rather than responding to the effects of the stolen election in Florida, or discussing institutionalised racism, these people might immediately respond by posting a link to a site which posits that black people are less intelligent than white people, but describe this site as “an analysis of the issue” – which is not true: the site is a distraction from the issue. As you are, it appears, the best example of “these people”, can you explain why you felt it necessary to bring up the nonsense claim of a “meaningful racial IQ gap” in a discussion about racism and the election in Florida?
Not that I really want to get into this, but I see Glaivester saying (in some places, at least) not that black people are too dumb to vote properly, and that if we studied the Florida election counts we’d see this, so that’s why no one wanted to look at spoiled ballots, but the very different claim that:
Even if provided with a history like the one provided here (that spoiled ballots are the result of people with a history of being discriminated against overmarking their ballots), most Americans, due to their own prejudices, would hear “Black people are too stupid to vote properly”, and no politician wants to be the one to have his views misrepresented that way. Consequently, Democrats who might have been sympathetic otherwise to counting the overvote, decided to let it go because they didn’t want to be in the inenviable position of looking as though they called a key constituency too dumb to vote.
That doesn’t strike me as a racist view, since it doesn’t seem to attribute stupidity to any group; but it does seem to me to be cynical.
True, journals do not prohibit somebody form publishing due to alleged racial views. However, people are not allowed to publish solely because they support the other side. Whatever side is making whatever claim in a journal, it has to satisfy an evidentiary, peer-reviewed standard to be even worthy of consideration of a place in the dispute.
When Jensen is accused of being a racist, it is said that his racist views predetermine his conclusions and that his conclusions of cognitive or intelligence differences between groups are founded upon his prejudice, not upon credible evidence or research. If you are saying that somebody is a racist because he/she believes research that is contestable but considered scientifically legitimate for debate, well, that’s another thing entirely.
Jensen is not a flaming racist unless his racial views are definitively the cause of his conclusions of potential cognitive or intelligence differences between groups. If that is such a blatant conclusion, then the APA has egregiously violated the evidentiary rules of peer-review, or they must think that his conclusions still may have basis in scientifically credible methodology.
Cala: Even if provided with a history like the one provided here (that spoiled ballots are the result of people with a history of being discriminated against overmarking their ballots), most Americans, due to their own prejudices, would hear “Black people are too stupid to vote properly”, and no politician wants to be the one to have his views misrepresented that way.
That isn’t what Glaivester said in his first comment on the issue: nor what the “analysis” he linked to was saying. G’s initial reaction was to say “Black people are too stupid to vote properly”, and to link to another site saying that again. Since G is available to explain why this was his first and foremost reaction, I figured he might as well do so.
Bunkum.
So long as Jensen’s data is valid and a genuine addition to what is already known, he will be published. Peer review determines the rigor of the data collection and basic analysis of correlations, that is all. That is an entirely different matter to whether his conclusions based on those data are racially biassed. Peer review is historically far less interested in conclusions than methodology – if one’s method is sound other people can build on that work no matter how eccentric the researcher’s conclusions might be.
Jensen is not being singled out merely because his views are politically incorrect. The further peer review of studies after publication, with accusations of overlooking confounding variables and underlying bias informing the interpretation of data, is just as common in journals concentrating on the tensile strength of metal alloys as it is in the social sciences. It just doesn’t get the publicity.
Cala:
I’m sort of saying what you said in the second paragraph, but I am also saying something similar to the first paragraph. I think that there is a high probability that on average people of African descent do have less of certain cognitive abilities (put another way, the IQ gap does measure something real). While some of that may be due to environmental factors (poverty, poor nutrition, perhaps vitamin D deficiency), I think that there is a fairly good chance that some of this is due to genetic differences.
Of course, there is also some evidence that people of African descent have advantages (on average) in improvisation and real-time thinking (part of which is reflected in their overrepresentation in fields such as athletics and jazz, both if which require the ability to improvise and to “think in real time”).
Having said this, I should point out that it is wrong to hold the “spoiled” ballots to a higher standard than that prescribed by law.
That is, it is regardless of why black people made more errors on their ballots, it is wrong to disqualify a ballot when the law says to count it.
However, even though the Democrats would be right to argue that many of the disqualified votes should have been counted, it would be somewhat embarrassing to admit that Democratic voters were more likely not to understand the ballot; and the race issue could not have been brought up; because while they would be accurate in arguing that according to the law, the standard required for a ballot to be counted should be lower than what was actually used in the recount, they couldn’t argue it without saying that the standards used needed to be lowered so that blacks could meet them, which doesn’t sound that good.
In short, the Democrats had the better legal argument, but it was impossible to bring it up without opening cans of worms.
As far as whether the disqualification was racist? I don’t think it was racist per se. It was simply partisan. The GOP held the ballots to a higher standard because it benefitted the party. They were happy with the disproportionate disenfranchisement of blacks not becasue thet disliked blacks per se, but because they knew that the blacks were most likely voting for the other guy.
Hmm, that’s not what the APA says its policy is during peer review. When it mentions
it includes among them,
Go visit this site.
Are you saying that confounding variables sometimes escapes the notice of peer review or that peer review will deliberately ignore confounding variables and let post-publication criticism deal with it? If you are suggesting the later, the APA denies allowing this in their peer review process, as confounding variables is grounds for outright rejection…
The APA manual further insists that submitted articles,
Go visit this site.
Disregard or adherence to the admonitions
They do not appear to be agnostic on the eccentricity or basis of the conclusion…
Go visit this site.
Allowing eccentric or unfounded conclusions not supported by the article’s data does not appear to be a policy of APA peer review (and hopefully not any other journal’s peer review). If Jensen’s conclusions are obviously founded upon his pre-existing racism and are not conceivably supported by the peer approved data and methodology in his submitted articles, the APA peer review must be chronically incompetent to keep allowing him to publish when his conclusions’ basis is his prejudice and not legitimate data. Maybe Jensen is bribing them.
I misspoke. Gould’s alleged debunking of factor analysis as a methodology for detecting g.
Sorry but I kept remembering when Gould called factor analysis a “bitch” and then said,
Since a unilinear scale of IQ is exactly one of the things that Gould was arguing against as invalid, and since he seems to be saying factor analysis inherently demands or justifies such a notion that he was criticizing factor analysis writ large as an intially flawed and prejudiced concept established for the express purpose of reifying fictional concepts, as when he said,
Since Gould finds many of those “particular theories of intelligence” to be canard and since he alleges factor analysis was established to bolster them, I kept recalling (incorrectly, I guess) that Gould found factor analysis to be partially illegitimate in itself. I thought that the fact that factor analysis is used to study non-socially constructed aspects was relevant.
Glaivester: As far as whether the disqualification was racist? I don’t think it was racist per se. It was simply partisan. The GOP held the ballots to a higher standard because it benefitted the party. They were happy with the disproportionate disenfranchisement of blacks not becasue thet disliked blacks per se, but because they knew that the blacks were most likely voting for the other guy.
I think that given you’re racist enough to believe that “blacks are dumb” (to summarise your comment) your assessment of whether the Republicans in Florida and elsewhere are racist must be considered unreliable.
However, I think it worth pointing out that a good reason why black voters in Florida and elsewhere tend to vote against Republicans is because Republicans are traditionally and still the party which is friendly to racists and to racist policies – one such is the deliberate disenfranchisement of black voters in Florida in order to give the illusion that George W. Bush had won the election. The Republicans are in a double-bind of their own making: racism is endemic in their party, and they are unlikely to attract many black voters while this is so. And so long as their reaction to most black voters preferring any other party is to disenfranchise as many black voters as they can, they cannot possibly either deserve or win any substantial fraction of the black vote.
Dorset writes:
Um, as I recall it, one of Gould’s objections was that factor analysis could just as well be used to analyze intelligence into two factors; and to this effect, he presented somebody’s alternative two-factor intelligence measure as an example of the arbitrariness of using factor analysis to claim that intelligence is accounted for by one factor.
His claims, as I recall them, are that you need an actual, substantial theory of what intelligence is in order to be able to decide how many factors you would need, that factor analysis can’t provide such a theory, and that the people applying factor analysis to arrive at one-factor measures of intelligence aren’t providing any such theories that pass muster, and try to wave the whole issue away by waving formulas around.
I thought the interview was very interesting, and the book sounds as if it should be a good (if scary) read.
There are two things I would really like to see about the 2000 election, though: 1) a report similar to the Washington Post-Miami Herald recount report, but more comprehensive. When you go through and evaluate every single ballot and categorize them so you can tally different ways (i.e., unequivocally for Gore, unequivocally for Bush, etc.), what was the vote total, depending on how you tally the categories? I don’t remember offhand if the Post-Herald recount report covered the discounted double votes or not, or if they covered only the hanging chad ballots. So I guess what I would like is a definitive, independent, reliable report that says basically, here are the vote totals if you count them this way, here are the vote totals if you count them another way, etc. Then we can actually have some hard numbers to discuss “who really won” Florida with. And 2) a definitive, independent, reliable report on all of the problems that were documented in the 2000 election and whether or not they were fixed for the 2004 election, and if not why not, and if so did the fixes work or not. All I have seen so far, for the most part, are either partisan broadbrushes or independent fine-details.
Pre-publication peer review is the beginning of a process, not the end of it. Of course papers with flaws pass publication review every so often, and if certain reviewers continually miss errors the readership pounces upon with glee, the editor stops using them. But I don’t claim that in this case the reviewers missed anything glaringly obvious. Confounding variables are not the end of the story, as per your quote from the APA above.
Criticisms of Jensen’s APA paper focus on whether IQ is a reliable measure of intelligence.
The APA has many members who make their living applying IQ tests, MMPI-derived tests, etc. There is a vocal minority within the APA and a larger body of critics outside the APA who argue that IQ tests are fundamentally flawed as objective measures of intelligence, but if the board which publishes the APA journal supports IQ testing as a valid analytical procedure, they can hardly dismiss a paper which relies upon IQ tests for the core of its data.
One can be wrong about a great many things without being in any way unethical.
Why isn’t the overvote racial difference due to the larger proportion of first time voters in the majority-black districts, where very vigorous voter registration and GOTV had proven to be successful? Along with the mentoring effect of going to the polls with an experienced voter to answer your questions (since the poll judges look very busy, and some people might tend to wing it rather than hold up a long line by asking a question to which they are 90% sure they know the (wrong) answer.
The Goopers look for ways to disqualify ballots disproportionately in majority-black districts or disqualify voter registrations in urban or majority-black districts since the GOP benefits from low turnout in those areas. Remember the Ohio GOP Secy of State made voter registration only valid if the voter registration cards were only heavy high quality paper stock, making it difficult for people to register anywhere else but at state offices, and making outreach efforts to register, perfectly legal in the state, very difficult in practice. This lasted about 2 days before the judge threw it out as patently violating Voting Rights Act. If rural whites were disproportionately under-registered or were first-time voters making the overvote, the GOP would move Heaven and Earth to make sure these folks faced no obstacles in registration or that their overvote would be counted according to their intention.
The IQ test is as much a social construct to promote White supremacy and male supremacy as the color coded terrorist warning system is a construct to manipulate the fear levels of the people in this country.
The tests aren’t flawed in the least bit, as they are doing exactly what they were intended to do: *prove* that White people are innately smarter than non-White people and that men are inately smarter than women.
But you know what? I did see one occasion where there were allowances given for the fact that Black people score lower than White people due to nonbiological or inate reasons, and that was during the “mental retardation”(which is what it’s called in legal terms, since the Atkins decision though they could have chosen better words) phase of a death penalty trial with a defendent who scored in the mid-sixties on a Worchester test with 70 being the cut-off score. In his argument that the defendant wasn’t “mentally retarded” and thus unable to be executed, he said that he should be given 14 points in his score due to his race and discrimination involving the IQ test, putting his IQ well above the cutoff score on the “death” side.
So don’t ever say that there aren’t people out there who are concerned about social/cultural/educational influences on IQ scores among different races. ***roll eyes***
Radfem: “The tests aren’t flawed in the least bit, as they are doing exactly what they were intended to do: *prove* that White people are innately smarter than non-White people and that men are inately smarter than women. ”
That’s curious since I believe (and correct me if I am wrong) that Asians score higher and as high as whites and has been shown elsewhere women, on average, score about the same as men, but there seems to be more men on the edges (scoring both higher and lower).
I am not sold that IQ tests are a reliable measure of how “smart” someone is, but I think they might be an indicator.
As an example Richard Feynman (a brilliant physicist who stood out among a group very smart people) only had a tested IQ of 125. Respectable number, but it in no way would have been a predictor of how “smart” that man was.
All that said, some ARE smarter than other people. If someone scores a 60 on an IQ test is there really any doubt in anyone’s mind that that person, what ever their racial make up, is mentally handicapped (or at least not as bright as most people)? Intelligence tests do measure SOMETHING don’t they? Being able to pick out the next logical item in a sequence doesn’t seem particularly biased to me. Surely such an ability in some way is a reflection of how smart someone might be. To what extent, I don’t think anyone really knows.
Actually, next logical item in a sequence is culture (and education) bound. As one of my highschool math teachers was fond of pointing out, if you are given a 3 number sequence, and asked for the next number, there may be an obvious answer, but there is pretty much never a wrong answer. For any 3 number sequence, you can construct a rule to give you any number as your next value. What sorts of rules are considered obvious (or even considered legitimate) is partly cultural, and purely learned.
On the other hand, being able to explain your rule by which you chose the next thing makes it less culture bound, except that it then adds on a layer of culture and education bound explaining skills, and the opportunity for prejudice on the part of the evaluator. Also, of course, if there actually is a cultural difference, then a reason that makes sense in my culture may seem completely screwy in yours.
Also, I think the question of whether IQ tests kind of measure something completely misses radfem’s point on whether they are used as a racist, sexist hammer. So long as they aren’t perfect measures of some innate quality, they are sufficiently flexible to be used as a hammer (hammers need to be flexible?).
Charles: “Actually, next logical item in a sequence is culture (and education) bound. As one of my highschool math teachers was fond of pointing out, if you are given a 3 number sequence, and asked for the next number, there may be an obvious answer, but there is pretty much never a wrong answer.”
Could try to explain that and maybe give some examples of culturally biased logic questions? In my opinion, understanding the question in context (very important) is an indicator of intelligence and is the first step to a reasonable answer.
Sure, if given the sequence 1/2; 1/4; 1/16 on some test and asked to give an answer for the next logical unit I suppose I could draw a picture of an “Apple” and explain that “well, that’s the how many pieces are there when I cut each piece of my apples in half in each step.” Sorry, no matter how many smiley face stickers you get on a math test such an approach would be incorrect in the context of the question.
But if instead someone answers 1/32, that would not be any less correct then 1/256. The question was ambiguous. Though more than one answer is correct, there are certainly wrong answers in the context of the question and part of intelligence is understanding that no matter your cultural background.
I read somewhere in some book that some kid was asked on a test to explain the steps he could take to use a barometer to measure the height of a building. If I remember this correctly he replied something like: “throw it off the top of the building and insert the time it takes to hit the ground in to this formula “. Then the teacher confronted him as told him that’s not the answer he wanted and gave the kid the opportunity to try again. The kid replied: “Set the instrument on the ground and make a mark at the top of the instrument. Then raise the bottom of the instrument to the line and make another mark at the top of the instrument again. Keep going up the side of the building until you get to the top…” Again confronted by the teacher the kid had several other novel ways to use the instrument to measure the height of the building (such as tying a string to the barometer, lowering in down until it touched the ground then measuring the string, etc). The point is, although the question was a little ambiguous, the kid was certainly bright enough to know the answer that teacher was looking for but purposefully chose not to.
Now maybe there are some cultural influences that might come up on the language portions of tests, but I have yet to see that demonstrated in a clear and convincing way (assuming of course that the test taker and the test use the same basic language). Otherwise it would seem that Asians, Indians (not native Americans), and other distinctly different cultures that routinely score high on IQ tests would suffer from some cultural bias as well.
Now maybe there are some cultural influences that might come up on the language portions of tests, but I have yet to see that demonstrated in a clear and convincing way
Sample question from IQ test: what’s the next word in this sequence?
Madison, Monroe, Adams, ?
Convincing enough for you?
“Convincing enough for you?”
Honestly, not even a little convincing.
First, it doesn’t seem like the kind of question on an intelligence test. Knowing Jackson was the next president (yes I had to google) is more about what you have learned in history class rather than something you can figure out on the fly using logic or reading comprehension (which is something IQ tests generally try to do). If that question was on an actual modern IQ test I would say its a lousy question.
Secondly, even assuming it was an actual IQ test question, would Asians somehow have a cultural advantage over whites and blacks about knowing the order of US presidents?
But understanding the context and intent of the question is entirely a matter of culture. Someone raised on logical sequence problems as a family game knows intuitively what is being asked for. Someone who has never seen a logical sequence problem before may well give the apple answer. They saw how the number sequence was strucutred, but they didn’t know how to frame the answer. It sort of like someone playing Jepordy who has never watched the show before, “What do you mean, ‘phrase the answer as a question,’ what the hell kind of sense does that make?”
Just because you are used to the way that logical sequence questions are structured (don’t tell me the rule of the sequence, or something that the sequence reminds you of, tell me the next number in the sequence using the most likely logical sequence rules) doesn’t mean it isn’t entirely culture bound. If you took a IQ test in a culture that valued the ability to construct koan-like metaphysial responses (something that smarter people will be able to do better than stupid people, for some meaningful sense of smart and stupid) we would both probabaly test out as seriously mentally deficient. And, if corelation with life outcomes is the best proof for the valididty of IQ, I’d point out that we’d also probably both be failures in such a culture. Would that mean we weren’t smart?
That the example you give as proof of it not being culture laden is so totally culture laden is certainly proof of something.
Also, you are still ignoring the fact that unless the test is perfect, radfem’s argument from how it is actually used still stands, even if it also serves as a passable test for some sense of intelligence.
Ability to function in the majority culture is a legitimate component of intelligence, just as ability to function in/relate to subcultures is a legitimate component of intelligence. Even more valid is the ability to on-the-fly assess which cultural expectation is appropriate in a particular context.
The genuinely-held belief that a picture of an apple is the expected answer to a numerical sequence is indicative of a profound cultural skills deficit in an individual – whether that deficit lies in unfamiliarity with mathematical logic, or whether the deficit lies in the ability to recognize whether a mental set based on logic or a mental set based on experience is the appropriate set to employ. Whether it is “fair” to measure or score that deficit is not particularly material; the deficit exists, and will correlate strongly to an individual’s ability to function intellectually.
Which, every question on an IQ test should.
IQ tests are horrible indicators of intelligence, even of the sort of intelligence for which they test. I know a person who came up with an 80 on one test and, less than a year later, scored 120. If there is that great of a range in less than a year in an adult, the testing method simply isn’t adequate. The other major problem with IQ as a measure of intelligence is that it is held (more in the past than now) as the measure of all intelligence (math, logic, social, cultural, etc) when it is clearly not.
Except that this isn’t what IQ tests are claimed to test. You can take the fact that the do also test that (okay, only the dominant culture part) and try to claim it is a positive, so if lower average scores for blacks as a group on IQ scores are a valid reflection of active racism on the part of test designers and givers, the lower test scores are accurate in so far as IQ is now a measure of how well you can (and that includes ‘are allowed to’) function in the dominant culture. The same racist bias of the IQ designers and proctors (my understanding is that many IQ tests are more heavily proctored than simple multiple choice or fill in the blank tests) is also present in the general White population, so the lowered IQ scores would accurately correlate to life outcomes (if your IQ scores are the victim of bias, so will be ther rest of your life), but to then call it intelligence seems dubious. “I’m just glad I was smart enough to be born into the dominant group.” Nu-uh.
If a particular group of people do badly on a particular test, that either means that the test is biased against that group in some way, or it means that what that test measures is something that group is deficient in. However, other than measuring people’s ability to take a specific test, tests are not perfect instruments, so if a particular group of people, who you have no other reason to think are deficient in the characteristic that the test is attempting to measure, get poor results on the test, then you can conclude that what the test is actually measuring is contaminated with something that that group of people are deficient in. Thus, the koan based IQ test, which does well for measuring the intelligence of people who have been raised working with and thinking about koans, will show Americans to be not very intelligent. Now, if one considers Americans to be abnormally stupid, then one will trumpet the koan IQ results as proof of one’s assumptions. However, if one doesn’t see any reason to think that Americans are particularly stupid, then one might suspect that there are cultural biases built into the koan IQ test.
Just for the record, I don’t think IQ tests measure “intelligence”. Intelligence is a very complicated thing; certainly too complicated to be reified into one number. IQ tests measure a certain subset of cognitive skills. That subset is a pretty darn important plotlinhe, but it isn’t the whole picture show.
The difficulty with your koan-based IQ test example is that presumes a separate culture which is not accessible to Americans. (Which is largely true – while individual Americans who are motivated to seek out knowledge of koans are capable of doing so, the number who are so motivated is very small, and the rewards for doing so not obvious.) Yeah, Americans would do badly on a test culturally themed around a Korean idea that we don’t use much. But so what?
In the case of testing within our own national group, there is no such separate cultural matrix. White Appalachian Catholic children are exposed to the concepts of numerical progression and time management, as are black Methodists in Texas, Asian Presbyterians in Minnesota, and Jewish agnostics in California. WASP kids in Connecticut aren’t being taught secret knowledge denied to 1st generation Sicilian immigrants; they’re just listening to the same knowledge better.
The old racial theories about IQ can be dismissed. The differences in group IQ performance are predicated on (and predictable by) the local immersive culture that a youth is exposed to. Black kids whose peers torment them for acting white if they come to class on time score worse on IQ tests than do black kids from certain neighborhoods in NYC whose parents beat the hell out of them for getting to class late. Groups whose immersive local culture – the group of peers, parents, role models and teachers – inculcate intellectual skills do well, and those whose immersive cultures don’t, do poorly.
The Flynn effect dovetails neatly with this phenomenon, as it reflects that by and large, once freed from gross discrimination, each generation puts more emphasis on education and learning than their parents did, because people can empirically see that doing so produces better outcomes for kids, and people love their kids and want them to succeed. Generally there is only so much improvement that can be done in a generation; my great-grandparents were dirt farmers, and by God they saw to it that Nona learned to read. Nona was a shopkeeper, and by God she saw to it that my mom got the chance to go to college. My mom was a schoolteacher, and by God she made sure that I went to the best school I could handle. Betcha a million bucks that IQ testing would show the Flynn pattern of improvements in our family in each subsequent generation. I would way outscore Nona on an IQ test – but she’s a shrewd old bird, and while I’m somewhat intrinsically smarter than she is, it’s not nearly as great a difference as the tests would show. I just got raised a lot better, mentally speaking.
Everybody can read Shakespeare and everybody learns to tell time. Some kids get ego reinforcement for reading Shakespeare and being prompt, and some kids don’t. That is sufficient to explain the group differences, from what I can see. It’s regrettable – tragic even – but it doesn’t make the tests invalid or worthless. Adults who as kids were trained to ignore time boundaries, avoid books, and devalue abstract knowledge are less competitive in the arenas considered important by predictive tracking (what we correlate IQ tests too) – the symbolic economy, and its educational parallel universe.
The tests do have cultural bias – they are biased against people who haven’t adequately assimilated their culture. That failure may well not be their fault, but it reflects a genuine difference in functional ability. In your koan example, the Americans who do poorly on the koan test ARE stupid – if they’re in a country where being able to manipulate koans is of utility.
I know a person who came up with an 80 on one test and, less than a year later, scored 120. If there is that great of a range in less than a year in an adult, the testing method simply isn’t adequate.
Yeah, that would be my assumption too. Unless your friend kicked the smack or something in the timeframe, then the testing method was poor.
Which means that the testing method was poor; it doesn’t say anything about IQ testing in general. It says that one (or both) of his testers fucked up, or that he himself was screwing around, or some other confounding variable was operative.
The best empirical research on “acting white” penalties for high grades is a paper a couple of Harvard folks published last year (pdf link). While those authors concluded that “acting white” does exist, the data doesn’t exactly match up with what you’re describing.
* For black students, the higher their GPA – up to 3.5 – the more Black friends they have. It’s only after GPA gets above 3.5 that the number of Black friends decreases. But a 3.5 is actually a very good GPA by most standards; so it’s hardly true – as most pop writings about “acting black” have it – that just cracking a book open or getting good grades will cause black students to be “tormented” by their black peers.
So you could just as easily claim, looking at the evidence, that black peer culture rewards good grades for all but the very top of the class, and that even for those making a 4.0, they are still – although less well-regarded than those making a 3.5 – better regarded than peers who get poor grades.
* The Harvard folks were only able to find measurable effects of “acting white” problems in public schools in which blacks were a 20% or less minority. In 80% and up black public schools, there is no “acting white” penalty associated with good grades – at least, none that was able to be measured. In all-black schools, the most popular students are the ones with the best grades. So if “acting white” penalities cause lower IQ scores, why don’t blacks students at mostly-black schools have markedly higher IQ scores than black students at mostly-white schools?
* High-achieving black girls had significantly less popularity drop-off than high-achieving black boys.
* Oddly enough, white private school students become less popular with their white peers once their GPA gets above 2.0. (The popularity of blacks at private schools is not related to their grades). What this means, I don’t know, but it indicates that “acting white” penalties are not, by themselves, capable of explaining the complex ways race, GPA and popularity interact.
* The “acting white” penalty problem appears much more severe among Hispanic students than among Black students.
* * *
There are, I have no doubt, multiple causes for lower Black IQ scores on average. One of those causes – and one that has FAR more empirical support than the “acting white penalty’ hypothosis – is “stereotype threat.”
Here’s Claude Steele describing a typical study aimed at measuring stereotype threat:
Robert,
It sounds to me that what you are saying is that cultural bias doesn’t have much to do with IQ tests, but in with intellectual pursuits in general. It sounds like you think the questions are not structured in a way as to be culturally biased at all. If I understand your position, that seems more likely, but…
I admit I am no expert in the area, but it seems to me that testing “intelligence”, if that is the goal, shouldn’t depend on a lot of factual knowledge gained in school (of course beyond a minimal level like learning the language of the test and basic math skills, etc), but the ability to accept, understand, process, and apply information given basic principles. Does that ability change much from a 13 year old child to an adult? If not, and it might be my ignorance in this area, your Nona shouldn’t be hindered by her lack of education on an IQ test.
As I alluded to in an above post, I don’t think IQ tests tell everything about how smart someone is. I think they certainly show important parts of the whole picture. Again, if someone consistently scores around 60 on an IQ test, is there going to be much doubt in your mind that that person probably isn’t as bright as most people, in general?
* A curious aside: I have always been curious about what Newton (imho one of the smartest people in recorded history) would have been like if he had been raised by nomadic Eskimos in relative isolation. Would we even know his name? Certainly there is the Indian example of another great natural genius Ramanujan. But while he was to some extent self-taught he wasn’t raised in complete isolation from textbooks and schools.
Amazing how the thread has drifted, despite efforts at correction, from the topic of overvotes, and by extension the mechanics of the voting process and counting of votes, in the FL election, to the old “race and IQ” rehash of the Bell Curve, and for that matter, perhaps even older eugenics screeds about how intellectually deficient the East European-origin Jewish immigrants were because they scored lower than native-born US AngloSaxon whites in IQ tests (that’s a 1910-1920 argument).
However, at the risk of further thread drift, I have to comment on Amp’s summary in #42:
“Oddly enough, white private school students become less popular with their white peers once their GPA gets above 2.0. (The popularity of blacks at private schools is not related to their grades).”
I wonder if the GPA of 2.0 is correct or a misprint, Amp. However, I can vouch for the inverse correlation of very high grades and popularity for white private school students, having been one myself. Geeks, especially female geeks, are viewed as weirdos and as social liabilities. Ever has it been thus, at least in the US. That’s why I don’t pay much attention to the blaming of blacks for the “acting white” stigma – the general culture, both black and white, reveres the jock, not the egghead.
I think that given Glaivester’s initial comment, the thread wasn’t so much “drifted” as “sabotaged”. :-)
Pingback: Alas, a blog » Blog Archive » Debate: It’s Time To Panic vs. There’s Light At The End Of This Tunnel
Pingback: Alas, a blog » Blog Archive » Rolling Stone Article: Was the 2004 Election Stolen?