Re: Education research exposes the theory of multiple intelligences as singularly stupid

— — —
I would separate the theory of different intelligences from the theory of different learning styles.I think they aren’t necessarily referring to the exact same thing. Also, conflating them probably isn’t helpful.
First, are there different intelligences? 
It would seem to be commonsense that some people are better (i.e., more ‘intelligent’) at doing certain activities. Even research confirms this. People can be tested separately for verbal intelligence, for mathematical intelligence, and for spatial intelligence. They have tests designed for measuring these abilities and they use them in schools.
However, this doesn’t necessarily have much to do with traditional teaching or necessarily should it. How is someone being good at sports going to help them learn to read? It’s not. A student may have natural verbal intelligence and not have natural mathematical intelligence, but that kid is still going to have to learn to do math and his verbal intelligence may not help him in any direct way.
The theory that people have different abilities is true, but the theory that these different abilities are different ways of learning isn’t proven, or not in all cases anyway. An ability can’t necessarily be applied as a learning style to learn other abilities.

Second, are there different learning styles? 

I was wondering what research has been done on this area. It’s not something I’ve looked into. However, I have spent years looking at the research on personality types and traits. I think it is a fair assessment to say that there is plenty of research supporting the theory that human brains work differently to varying degrees. There are commonalities to human brain function, but there is also much diversity. It seems obvious to me, going by the research I’ve seen and going by my own observations, that people do process information differently. This potentially could lead to different learning styles, whether or not such learning styles correlates to Howard Gardner’s theory of multiple intelligences.

This is an areas I can speak personally about. I had a learning disability as a child, still do. My learning disability just meant I learned differently, but it didn’t mean I was less smart. When I was in early elementary school, I wasn’t keeping up with the other students. They, at first, suspected I might be low IQ. But when they tested me, they found I was actually above average in IQ, at least in certain areas. For example, my ability to solve spatial puzzles was at a twelfth grade level.

So, that is my learning style. I’m very good at seeing patterns and making connections. However, I’m not good a rote memory. Traditional teaching methods didn’t work for me. Fortunately, I went to school that had a very good special education teacher who taught me a different way to learn. A main part of my learning disability is word recall. For this reason, I was behind on my reading and writing ability. One thing the special education teacher taught me was to think of similar words. If I can’t think of one word, I use another similar word. This caused me to increase my vocabulary and after that I was ahead of the other students in reading.

In my case, there was some connection between intellectual abilities and learning styles. If you tested me as a kid, I would have measured as low in verbal intelligence. But if you test me now, I would measure as high in verbal intelligence. So, what is the difference? I learned to use my other strengths (i.e., my other intellectual abilities) to compensate for my weaknesses (i.e., my learning disability, especially word recall). It would seem that word recall is a part of verbal intelligence. It would seem that my limited word recall ability means I lack a certain natural aspect of verbal intelligence. However, because I was intelligent in other ways, I could compensate. People who don’t have other types of intelligence wouldn’t be able to easily compensate. Even I, if I hadn’t had a good special edcuation teacher and parents who were teachers, might not have learned to compensate. If I had been a poor kid in a poor public school, I might have been simply categorized as low IQ and that would have been the end of it. For the rest of my life, I would have thought I was stupid.

As such, I’m a strong proponent of learning styles theory. I wish I had the opportunity to be taught differently all through school. After leaving that particular elementary school, I never again had a teacher that helped me to that extent. Despite being above average in intelligence, I always struggled with learning and so I learned to hate school. Teachers were always teaching rote memory, but I was never good at it. I dropped out of college partly because no one would teach the way I learned. The way I learn is by connecting information, by seeing or creating patterns in data. Isolated factoids are meaningless to me and it’s hard for me to remember them even if only for the short term of taking a test.

I was, however, lucky to have parents who were teachers and very helpful in my education. My mom was a speech pathologist and so she was used to kids who had minds that worked differently. So, despite school, my parents instilled in me a love of learning. I learned how to learn on my own.

Not everyone is a genius, that is true. On the other hand, not everyone who learns differently is stupid.

— 
 — — — 

Even so, I think the research supports the theory of general intelligence. 
If I wasn’t generally intelligent, I wouldn’t have been able to compensate for my learning disability. General intelligence would be the measurement across all testable cognitive abilities: verbal, mathematical, spatial, etc. I was thinking about the research showing that people who learn music early are better at math later on. So, certain aspects of intelligence would probably be more closely correlated than other aspects. I might be above average on general intelligence or maybe not, but I have no particular talent for either music or math. That side of general intelligence is separate, in my personal experience, from verbal and spatial intelligence. But I would guess that, for most people, there is a correlation between verbal intelligence and mathematical intelligence.
Verbal intelligence is, as I understand, often used as a stand in for general intelligence. In traditional education, verbal intelligence is key since almost everything is taught through language. However, there is no reason that needs to be the case. There are some autistic kids who have limited verbal intelligence and yet have vast mathematical intelligence. So, general intelligence is only a generalization and wouldn’t apply to all people. Even if general intelligence is the rule, schools still need to educate those who are exceptions to the rule.
Nonetheless, I see the risk in emphasizing differences over similarities.It fits our cultural ideology to believe that everyone is smart or talented in some way, that everyone has great potential if we could only find a way to tap into it. I’m sure there is much truth to this, but on a practical level it probably makes more sense to emphasize standardized education methods.

It can be an excuse to just dismiss general intelligence just because it is inconvenient to our ideology. I was thinking of two examples where this becomes problematic.

 – – –

The first example is differences in IQ between races.

Some people want to dismiss IQ testing because it has the potential to be used to support racist theories and those racist theories can influence public policy. But the problem isn’t the IQ tests. The problem I see is in understanding what the IQ tests represent.

IQ isn’t simply about natural ability. Kids who don’t get proper nutrition while the mother was pregnant and after they are born will tend to have lowered cognitive development. Kids who experience high rates of social stress and environmental toxins (especially lead poisoning) will tend to have lowered cognitive development. Kids who don’t have access to materials to stimulate their minds and don’t have parents with lots of free time to work on their learning will tend to have lowered cognitive ability. These types of kids are disproportionately found among the poor and minorities are disproportionately found among the poor. So, it’s no surprise that minorities on average have lower IQs when tested.

To demonstrate this, all you have to do is look at the IQ results as kids grow older. When young, there is no clear difference between the IQs of whites and minorities. The difference only shows up in later years of schooling. So, there is no evidence that the racial difference in IQ is genetic.

Because we see these racial differences, IQ testing is all that more important. It helps us to discover the causal factors in these racial differences. In understanding these causal factors, we can change them. We can take actions through public policy that increase the probability that poor children get proper nutrition, have less stressful environments, don’t get lead poisoning, and have intellectually stimulating materials and environments.

– – – 

The second example has to do with the IQ differences between liberals and conservatives.
Many different studies have found that liberals on average have higher IQs. There are many theories for why this is the case, but no one knows for sure the reasons.

Like the liberals who want to dismiss the racial IQ difference, there are conservatives who want to dismiss the ideological IQ difference. There are always people who want to dismiss inconvenient and uncomfortable data. I came across someone attempting to do that with the ideological IQ difference. Satoshi Kanazawa has proposed an explanation for why liberals test as more intelligent. In response, Shawn T. Smith has dismissed that explanation and in doing so seems to dismiss the idea of general intelligence as something IQ tests can measure or else he is just dismissing the correlation of verbal intelligence to general intelligence.

Here is an interaction that shows the conflict between the two views (from the comment section of Shawn T. Smith’s article):

RobertS (April 11, 2010 – 10:25pm): You ignored the studies Kanazawa referenced in his paper SOCIAL PSYCHOLOGY QUARTERLY “verbal intelligence is known to be highly correlated with (and thus heavily load on) general intelligence. Miner’s (1957) extensive review of 36 studies shows that the median correlation between vocabulary and general intelligence is .83.” He also cites Wolfle (1980) and Huang and Hauser 1998. My guess is your a conservative who’s offended by Kanazawa’s conclusions.
Shawn T. Smith, Psy.D. (April 12, 2010 – 8:07am): With all due respect, I absolutely did not ignore his citations. Of course verbal intelligence is heavily loaded on general intelligence. So is PIQ. This proves what?

Let me ask you: do you believe that extrapolating a Full Scale IQ score from a vocabulary test is a valid, reasonable, and accurate thing to do? Don’t tap dance. Give me a straight answer. If you do believe it is a valid thing to do, tell me why. Maybe I’ve got it all wrong.

Also, do you agree with Kanazawa that liberals are more intelligent than conservatives? If so, why?

Yes, I suppose I was personally offended, but not because I am conservative (incidentally, I described my stance briefly in my deconstruction so you would know where I stand). I’m becoming increasingly offended by the degree and amount of irrationality and intolerance in my profession, both in research and in the clinic. It does offend me to see research abused this way. It makes all of us psychologists look silly.

Here is another interaction about the same issue:

MP (November 8, 2010 – 8:16pm): I believe the original author showed a second order correlation between liberalism and verbal intelligence.

This takedown is really really nitpicky, the sort of nitpicky that that “they’re” vs “there” guy was doing earlier.

The correlation is extremely strong, so much as the data shown. The LSE guy might have overstated his case but the numbers don’t lie.

That crap about colleges = liberalism doesn’t weaken the correlation.

>>>>Second, a correlation once-removed is terribly shaky. What Kanazawa did here is akin to saying, “tables have four legs like dogs, and dogs resemble elephants in many ways, therefore tables are elephants.”

Errr no. I was taking you quite seriously until I saw this comment you made. I think your complaints are minor, but fair. But this statement is either stupid or disingenuous. He never claimed that liberals are smart. As we people who understand highschool statistics know, he’s saying there’s a correlation between intelligence and liberalism. That’s like saying a table is more similar to an elephant because both have 4 legs, than to a rock, which doesn’t.

PS: Your analogy sucks.

Shawn T. Smith, Psy.D. (November 8, 2010 – 9:56pm: You said that my criticisms are nitpicky, but you glided right past the most salient ones. For example, I criticized Dr. Kanazawa for extrapolating full scale IQ score ranges from a rudimentary vocabulary test. Do you believe that to be a trivial point?

And I stand by my criticism of second-order correlations. We in the field of psychology have become far too accepting of strained correlations, meta-analyses, and other manifestations sloppy research. You and I may differ on that point because I am a clinician and I deal in individual cases, not numbers. If you deal in statistics – not that it’s good or bad, just different – you will undoubtedly see more value in things like second-order correlations.

But back to my original question: Was Kanazawa correct in deriving full scale IQ ranges from the vocabulary test? (Yes, I realize it can be done statistically. Anything can be done statistically.)

And one more interaction:

Steven (February 1, 2011 – 4:02pm): I read the deconstruction at ironshrink.com primarily because I was shocked by the large difference and wanted to analyze the methodology. Unfortunately, what you gave was incredibly biased in its own right. I will just look at your main two premises:

1) PPVT is not a good measure of IQ

2) Liberal Colleges produce Liberals

The second one is completely illogical. I think we can agree that college graduates are smarter in general. Then you argue, of course college graduates (smarter people) are going to be more liberal, they are taught by liberals! Thus you are accepting the Hypothesis in order to disprove it.

Now the first one, which is an okay complaint, however, your emphasis on it is way too strong. Using PPVT and reporting IQ does damage the accuracy of the study, but to be fair, in order to make the study understandable to the general public IQ needs to be used. Furthermore, there have been studies that show that PPVT is a moderately good predictor of IQ, and you say it is often argued as a fairly good predictor of verbal intelligence.

Thus at best your ‘debunking’ shows that instead he should have written “Liberal are more verbally intelligent than conservatives” not merely intelligent.

Sorry, but your debunking is more biased than the study

Shawn T. Smith, Psy.D. (February 1, 2011 – 4:23pm): Good effort, but I’m not convinced.

1) PPVT is not even close to a measure of intelligence. I won’t restate my case here since I explained it in detail in the deconstruction to which you refer. Kanazawa’s attempt to connect the PPVT to full scale IQ is unsupportable. Had I tried a maneuver like that on my dissertation, I would have been laughed out of grad school. And rightfully so.

2) Evidence does not support the idea that college graduates are more intelligent than non-grads. Intelligence is a multi-faceted construct. College grads tend to do better on the verbal portion of IQ tests, and so it is no surprise that they gravitate toward an environment where verbal strength is an asset. That does not mean that college graduates function equally well in other areas of life requiring other skills. This explains why accountants and mechanics hire each other. They each play to their strengths – at least they do if they’re intelligent.

To summarize, Shawn T. Smith makes three related claims:

1) Standard methods of testing (in education, in psychological research, and in the legal field) isn’t to be trusted as a reliable and unbiased measure of general intelligence. Most studies confirm these standard methods of testing because most studies are biased in the same way. Since the studies of IQ can’t be trusted, the correlations determined by those studies can’t be trusted.
2) “Intelligence is a multi-faceted construct.” Hence, it is to be doubted that there is a singular general intelligence or, if it exists, that it can be measured by the standard methods of testing. Even if liberals were more verbally intelligent, it wouldn’t lead to the conclusion that liberals have a higher rate of general intelligence.
3) Colleges are biased toward both verbal intelligence and liberalism. As such, psychological researchers and the entire field of academic psychology is biased in the same way. Verbally intelligent, liberal psychological researchers are merely prejudiced against those who are different than them: conservatives and those who are intelligent in non-verbal ways.

 – – – 

Here is my response:

Almost all intelligence relates to verbal intelligence. Even a mechanic needs the ability to have good reading comprehension in reading highly technical manuals. If verbal intelligence correlates to liberalism, it would be expected that more successful mechanics would be more liberal on average than less successful mechanics.

Besides, college doesn’t just teach verbal intelligence. If conservatives have other types of intelligence, then they should succeed just fine in certain academic fields just as long as they have moderate levels of verbal intelligence.

I don’t understand why Smith wants to dismiss verbal intelligence. It would be hard to be highly successful in any type of job without at least average if not above average verbal intelligence. If conservatives are below average in verbal intelligence, you’d expect to find them to have below average representation on the successful end of almost all careers.

For example, owners of janitorial businesses are probably on average more verbally intelligent and more liberal than the janitors that work for him/her. A business owner needs verbal intelligence to understand all the complex laws and tax policies and needs verbal intelligence to know how to communicate well.

The correlation still remains strong and still remains relevant. All Shawn has demonstrated is that he doesn’t like the correlation. If he can prove the correlation is false, then he should do some peer-reviewed research that proves his hypothesis. Until then, his empty speculations are simply mindless complaining by an ideologue.

We must base our conclusions on the known facts. Numerous studies have confirmed the correlation Kanazawa has made. There are also conservative psychological researchers. If they disagree with Kanazawa, I’m sure the liberal psychological researchers would welcome quality peer-reviewed research to further understanding of this issue. To put it simply, put up or shut up.

I could counter Smith’s argument by pointing to all of the other studies showing a correlation between liberalism and IQ. For example:
Even Kanazawa mentions other data that corroborates his own study:
Kanazawa quotes from two surveys that support the hypothesis that liberals are more intelligent. One is the National Longitudinal Study of Adolescent Health, which is often called Add Health. The other is the General Social Survey (GSS). The Add Health study shows that the mean IQ of adolescents who identify themselves as “very liberal” is 106, compared with a mean IQ of 95 for those calling themselves “very conservative.” The Add Health study is huge — more than 20,000 kids — and this difference is highly statistically significant.
But is there any point in mentioning further data? Smith doesn’t trust the methods of these kinds of studies nor does he trust the people doing the studies. Once such mistrust is allowed, discussion becomes impossible. This is similar to the danger in the education system when multiple intelligence theory is used to dismiss standard teaching methods. A conservative like Smith can say all education is biased against conservatives because most teachers are liberal and most teaching methods favor verbal intelligence. There is some truth in this, but it’s not a very helpful criticism.
What is expected? Are schools supposed to have quotas to ensure there are equal number of conservative teachers? And are teachers supposed to stop using verbal teaching methods?
Maybe it would be better to simply try to understand the data we have rather than dismiss it or attack the messenger. If we follow down the pathway Smith points to, we would find we have lost any bearings amidst all of the distrust and paranoia. Instead, we could look for reasons why conservatism correlates to lower IQ in the same way we look for reasons why poverty correlates to lower IQ. For example, we could ask:
What do conservatives parents do in raising children that is different liberal parents?
One answer is that conservative parents are more supportive of spanking. Recent research has found there is correlation between spanking and lower IQ. It’s probably similar to aspects of poverty. Spanking causes a child stress. Research also shows social stress in general correlates to lower IQ. So, conservative parents might want to rethink the conservative practice of spanking, assuming they care about their children’s cognitive development.
— — —
Here is my conclusion:
My point is that a balanced approach is required for understanding such issues. We need to take the data seriously, even when it contradicts our favored beliefs and values and especially when there is consensus among experts and researchers. Also, we need to gather as much data as possible. Sometimes there is elements of truth in opposing viewpoints, and sometimes not.
Furthermore, I’d differentiate between knowledge and practice. It could be factually proven that multiple intelligences exist and yet that doesn’t prove that this is helpful in teaching. Or it could be factually disproven that multiple intelligences exist and yet that doesn’t disprove that different teaching styles/methods are helpful to different students. We need to base teaching on what is demonstrated to work rather than on theory of what should work.