Sunday, December 19, 2004
revenge of the nerd discussion
My post last week on the causes of nerddom seems to have touched a nerve, as evidenced by the relatively large response (relative to no one responding to my other posts). I thought I'd take a moment to respond to my critics, and... here's the exciting part... present some new evidence.
Now, to recap, the article that I was responding to argued that nerds are nerdy because they choose to be -- because they value smarts over popularity. I disagreed, arguing a) that most nerds do in fact desire popularity, and b) that nerds aren't actually smarter than non-nerds, they are just worse at hiding the intelligence they do have.
Jamie writes:
Liz Speck writes:
Again, I agree but I don't think this contradicts what I've said. First, I was basing my "personal experience" section on my three years at Dover Junior High, not on Exeter and Harvard, which I recognize are effete, dork-celebrating institutions. Second, my point wasn't that intelligence isn't stigmatized. It is, greatly. It's just that I think some smart kids do a good cover job (and get popular) while others are less successful (and get pantsed). And that the kids who do a bad job aren't any smarter than the kids who do a better job. They're probably more cued in socially, but maybe too they've got some countervaling positive (like being good at sports) that makes people more forgiving towards them, or else maybe they're just lucky.
While thinking about all this I suddenly realized that I was supposed to be an economist, and that I had at my disposal a large dataset (90,000 students) with lots of information and test scores and friendship and stuff. So here's what I did: I constructed a rough measure of popularity. Every student was asked to list their 10 best friends, so for each student I counted the number of people who had listed them. This varied from 0 (many people had no one list them) to a max of 37 for what must be one wicked cool kid. Then I saw how this popularity measure was related to grades and test scores. The results were surprising.
And increase of 1 on the popularity scale (an additional person listing you) was associated with a half letter-grade rise in GPA. And lest you think this is some weird affect of people behaving differently in class, where people can observe each other, it is also associated with a third of a standard deviation rise in standardized test scores, which presumably are private information. So the data suggests rather strongly that, contrary to common wisdom, popular kids are actually smarter than nerds, or at least do better in school. One caveat: this says nothing about casuality. We can't say that good grades lead to popularity or vice versa, only that people with more of one tend to have more of the other.
I confess to being surprised by this. At most I thought I'd find no correlation. Could common wisdom actually be wrong here? Could there be a small clutch of extremely bright nerds who set the stereotype, then a large mass of low achieving vaguely nerdy drones?
Now, to recap, the article that I was responding to argued that nerds are nerdy because they choose to be -- because they value smarts over popularity. I disagreed, arguing a) that most nerds do in fact desire popularity, and b) that nerds aren't actually smarter than non-nerds, they are just worse at hiding the intelligence they do have.
Jamie writes:
I think Jamie's right that there will always be social divisions in schools, and that some kids will be ostracized no matter how hard they try to avoid it. And that it isn't simply being hip to social cues -- it's looks, sports ability, wealth, luck, and a host of other things outside a person's control. However, I still think social adeptness will always top that list. And I will also point out that this comment reinforces argument (a), that nerds usually don't want to be nerds and use all their wits to find a way out. They just aren't always successful.I don't think it's fair to blame the fact that kids are unpopular or get picked on at school because "they just can't pick up on social cues". I spent most of middle school and the bulk of high school tormented by the fact that I didn't have any close friends, and the "friends" I did have mostly made fun of me because I sucked at pretty much all the things we did together: video games, basketball, tormenting those slightly less popular than us. I was surpressing my intelligence and eccentricities, desperately trying to figure the angles and was still getting nowhere... even if you had a whole middle school of eccentricity suppressing social cue Sherlock's it would still break down into in-groups, out-groups, and free floating kids who sit alone at the end of the lunch table.
Liz Speck writes:
I'd just like to point out that you hardly get an accurate cross-section of the
popularity castes from Exeter and Harvard. I think you're right to a certain
extent, but my public high school experience was wildly different from my
experience at Exeter. There most definitely was a stigma attached to
intelligence, unless you were letting those terrible bitches copy off your
grammar test. Grrrrr.
Again, I agree but I don't think this contradicts what I've said. First, I was basing my "personal experience" section on my three years at Dover Junior High, not on Exeter and Harvard, which I recognize are effete, dork-celebrating institutions. Second, my point wasn't that intelligence isn't stigmatized. It is, greatly. It's just that I think some smart kids do a good cover job (and get popular) while others are less successful (and get pantsed). And that the kids who do a bad job aren't any smarter than the kids who do a better job. They're probably more cued in socially, but maybe too they've got some countervaling positive (like being good at sports) that makes people more forgiving towards them, or else maybe they're just lucky.
While thinking about all this I suddenly realized that I was supposed to be an economist, and that I had at my disposal a large dataset (90,000 students) with lots of information and test scores and friendship and stuff. So here's what I did: I constructed a rough measure of popularity. Every student was asked to list their 10 best friends, so for each student I counted the number of people who had listed them. This varied from 0 (many people had no one list them) to a max of 37 for what must be one wicked cool kid. Then I saw how this popularity measure was related to grades and test scores. The results were surprising.
And increase of 1 on the popularity scale (an additional person listing you) was associated with a half letter-grade rise in GPA. And lest you think this is some weird affect of people behaving differently in class, where people can observe each other, it is also associated with a third of a standard deviation rise in standardized test scores, which presumably are private information. So the data suggests rather strongly that, contrary to common wisdom, popular kids are actually smarter than nerds, or at least do better in school. One caveat: this says nothing about casuality. We can't say that good grades lead to popularity or vice versa, only that people with more of one tend to have more of the other.
I confess to being surprised by this. At most I thought I'd find no correlation. Could common wisdom actually be wrong here? Could there be a small clutch of extremely bright nerds who set the stereotype, then a large mass of low achieving vaguely nerdy drones?
Comments:
<< Home
I actually seem to recall research that said exactly and precisely what you are saying (at least, it's what I think you are saying)--that there is a positive correlation between grades and popularity. I also seem to recall that those researchers were somewhat surprised at their results, too. But unfortunately I remember nothing else about the research, except that it was not done within the last few years, but some time ago.
Interesting. I would have expected no correlation too, but I guess it could be because kids happier at school are able to achieve more.
At my school there were certainly a range of abilities in each of the social strata. There were kids both smart and cool, and kids dumb and geeky. They deserve pity, more than anything, the people who do nothing but work their arses off for years but still fail to achieve. Myself, I was never one of the 'cool kids', but was happily somewhere in the middle of the rankings. High enough to have friends, and to be able to achieve at school.
I'm now reading Physics at Cambridge (the one in England) so I know exactly what you mean about institutions that positively celebrate nerdiness. The department is populated by people that simply couldn't survive in any normal walk of life. Makes you wonder what happens to the nerds who don't make the grade intellectually...
-Grumbledore
At my school there were certainly a range of abilities in each of the social strata. There were kids both smart and cool, and kids dumb and geeky. They deserve pity, more than anything, the people who do nothing but work their arses off for years but still fail to achieve. Myself, I was never one of the 'cool kids', but was happily somewhere in the middle of the rankings. High enough to have friends, and to be able to achieve at school.
I'm now reading Physics at Cambridge (the one in England) so I know exactly what you mean about institutions that positively celebrate nerdiness. The department is populated by people that simply couldn't survive in any normal walk of life. Makes you wonder what happens to the nerds who don't make the grade intellectually...
-Grumbledore
# posted by 
: 8:18 AM
: 8:18 AM
Ah, Nerds. It’s tough to pass up an opportunity to let out my opinions on them, even when I don’t have any data to back myself up. I have throw around a lot of terms to say anything about this, and just I hope that no one will take me too literally or get offended or anything. These are just my late-night thoughts on the subject.
It seems that a lot of the disagreement that I see in this discussion lies in the definition of the terms. How do you define nerds, popular kids, and intelligence? And can one explanation last for a lifetime? I have the sense that most people have not differentiated between middle and high school in their theories. I find, however, that at some point around high school, there is often a big change in what it means to be cool. Pretty most everyone wants to be popular around the beginning of middle school when popularity really kicks in as a concept, but by the end of high school a lot of people have become very content not to be popular.
One problem with saying that everyone is always trying really hard to be popular is that it offers no explanation for kids playing “Magic: The Gathering”, trying out for the academic olympics, or otherwise wearing indisputable badges of dorkdom and nerdia proudly on their chests. At some point, many kids really do decide to do things that they know hurt their case for popularity.
Why? Because what kids really want is some degree of acceptance into a social group more than simple popularity, and many people find that acceptance in uncool places. While the cool kids will definitely give you swirlies if they see you with your dungeon master’s guide, the other dorks will also come to comfort you afterwards, and will hang out with you and roll the twelve-sided die. So if you’re already a little scrawny and bookish, why not give up on the holy grail of adolescent status and instead just hunt for the orb of ganthar?
To simplify things, I think that the article was wrong to say that nerds are smart kids who don’t care about being popular, but I also think that Alex is wrong to imply that nerds are dumb kids who do. It seems to me that nerds are defined by the combination of a lack of social skills and an open interest in academic subjects. Were we to study them, I believe that we would need some kind of indicator outside of grades, test scores, or popularity, such as membership in nerdy clubs (computer club, math team, library a/v aids, etc).
Paul Graham doesn’t define nerds, but claims that “there is a strong correlation between being smart and being a nerd.” I’m not sure about this, but at least given my definition than I wouldn’t be surprised if he is wrong. Either way, if we intend to collect actually data, we need to separate the concept of “smart kids” from “nerds”. I would define “smart kids” as members of any social group who have high indicators of intelligence. Unfortunately, as we probably are not going to find any better indicators than test scores or grades, we might want to divide this category into “good students” and “good test takers” to be absolutely fair.
Trying to figure out which kids are “smartest” is difficult if you accept the idea (as I do) that there are multiple forms of intelligence. I don’t know how scientists tend to define them, but I imagine them to include language, mathematics, spatial analysis, social and emotional perceptiveness, and some sort of logic or analytical ability. Some of these, such as math and logic, tend to carry much more or a “nerdy” stigma. They are also viewed differently in boys and girls. Others, such as emotional and social intelligence, seem to have less relevance to school and a great deal of relevance to number of friends.
Even if you were to accept that good grades and test scores generally indicate some sort of overall smartness, there is still the problem that these measures are very much tied into the confusing web of popularity, self-confidence, and other indicators of adolescent school performance.
I wonder how popularity correlates with minority ethnic backgrounds and low family income, which are both indicators of lower grades. Kids for whom English is not a first language, for example, tend to get worse grades. But are they less popular? Perhaps in schools where most kids do speak English fluently, but what about schools with very large non-native English speaking populations? And are members of all of these groups less “smart”? I definitely wouldn’t say so. Accepting these indicators of intelligence definitely creates some issues.
Looking at popularity without any reference to race, class, or gender is also problematic. My own experiences indicate that class issues are pretty heavily intertwined with all of this. In my high school, pretty much all of the kids who were noticeably poor were not cool. But there was also some degree of stigma against the kids who were noticeably rich. The popular kids were, almost exclusively and in almost every sense, average.
The problem is that average means different things in different places. Of course we can look at national averages, but my theory is that part of popularity is being fairly average for the particular place where you are. This also raises the question of school size and other skewing factors. Do people in big schools tend to have more friends? There were only 21 kids in my grade in middle school, so I didn’t even have a shot at the 37 required to put me in first place. I wonder how school size correlates with academic performance?
In general, I have a lot of questions about the skewing potential. What would it mean if there is a correlation between the general academic performance of a school and the general level of popularity of it’s students? I could also believe that successful schools tend have less entirely friendless students.
Given all of this, I’m at the point that I often reach with the social sciences: Is this really going to get us anywhere? I am skeptical of any conclusion that we reach being anything more than a nominally educated guess. But, since I have come this far, I’ll try to continue to lay out definitions from which we could move forward.
Even if we accept it’s conceptual validity, popularity remains difficult to measure. In Alex’s mini-study, he goes strictly on numbers and determines that there is a positive correlation between number of friends and grades and test scores. It seems to me, however, that popularity is less how many friends you have and more who they are and what they do. I view “popular kids” as a specific social group, just like jocks or nerds or dorks or hicks or anything else. Popular kids were the kids who cared the most about being popular, and found success in meaningless popularity contests, like the “most popular” class superlative, homecoming queens, student council prom court, etc. At the same time, popular kids are also bonded by the exclusion of others. They affirm themselves and deal with their lack of self-confidence by putting down others. Popularity is not just picking up social cues, but using them to administer a beat down on all of those less able to grasp them.
This, however, is hardly unique to popular kids. It is the same as jocks looking down on everyone who sucks at sports, and kids who do well in school looking down on those who do poorly, and kids who can write computer programs looking down on those who struggle to open word processor documents. I believe that each group is formed and defined in large part by some sense of superiority over others. Again, this may shift over time. In early middle school it may be that the popular kids simply begin excluding those that are not popular. But somewhere along the way, it begins to be a more of a reciprocal relationship. Certainly not all groups of equal strength, and often nerdy groups are really excited if someone “popular” wants to hang out or join them in a way that “popular” groups would not treat a “nerd.”
If you accept that popularity is a group and not a number, than Alex’s findings make more sense. Within each group, the kids that do the best in school also tend to have the most friends. It seems likely that these kids also tend to be more outgoing, well-adjusted, self-confidence, and come from a more stable family situation. But this doesn’t mean that they’re smartest, or that they’re in with the in crowd. It just means that they’re pretty on top of things. At the same time, kids who have the least friends are by no means the nerdiest. They probably have much more going wrong in their lives than those kids getting the best grades, or are the victims of prejudice and policy that hurt both their grades and their social life. Difficult family situations definitely are problems both with school and with friends. It makes a lot of sense.
There are still a lot of questions. All of my definitions have the problem of tending to make things seem binary, when in fact these issues of much less black and white. I maintain my suspicion that popular kids are average, with a tendency against those who are extremely smart or dumb, as well as other prejudices. And I’ve gotten too distracted by all of the details of this to have other big opinions right now.
Post a Comment
It seems that a lot of the disagreement that I see in this discussion lies in the definition of the terms. How do you define nerds, popular kids, and intelligence? And can one explanation last for a lifetime? I have the sense that most people have not differentiated between middle and high school in their theories. I find, however, that at some point around high school, there is often a big change in what it means to be cool. Pretty most everyone wants to be popular around the beginning of middle school when popularity really kicks in as a concept, but by the end of high school a lot of people have become very content not to be popular.
One problem with saying that everyone is always trying really hard to be popular is that it offers no explanation for kids playing “Magic: The Gathering”, trying out for the academic olympics, or otherwise wearing indisputable badges of dorkdom and nerdia proudly on their chests. At some point, many kids really do decide to do things that they know hurt their case for popularity.
Why? Because what kids really want is some degree of acceptance into a social group more than simple popularity, and many people find that acceptance in uncool places. While the cool kids will definitely give you swirlies if they see you with your dungeon master’s guide, the other dorks will also come to comfort you afterwards, and will hang out with you and roll the twelve-sided die. So if you’re already a little scrawny and bookish, why not give up on the holy grail of adolescent status and instead just hunt for the orb of ganthar?
To simplify things, I think that the article was wrong to say that nerds are smart kids who don’t care about being popular, but I also think that Alex is wrong to imply that nerds are dumb kids who do. It seems to me that nerds are defined by the combination of a lack of social skills and an open interest in academic subjects. Were we to study them, I believe that we would need some kind of indicator outside of grades, test scores, or popularity, such as membership in nerdy clubs (computer club, math team, library a/v aids, etc).
Paul Graham doesn’t define nerds, but claims that “there is a strong correlation between being smart and being a nerd.” I’m not sure about this, but at least given my definition than I wouldn’t be surprised if he is wrong. Either way, if we intend to collect actually data, we need to separate the concept of “smart kids” from “nerds”. I would define “smart kids” as members of any social group who have high indicators of intelligence. Unfortunately, as we probably are not going to find any better indicators than test scores or grades, we might want to divide this category into “good students” and “good test takers” to be absolutely fair.
Trying to figure out which kids are “smartest” is difficult if you accept the idea (as I do) that there are multiple forms of intelligence. I don’t know how scientists tend to define them, but I imagine them to include language, mathematics, spatial analysis, social and emotional perceptiveness, and some sort of logic or analytical ability. Some of these, such as math and logic, tend to carry much more or a “nerdy” stigma. They are also viewed differently in boys and girls. Others, such as emotional and social intelligence, seem to have less relevance to school and a great deal of relevance to number of friends.
Even if you were to accept that good grades and test scores generally indicate some sort of overall smartness, there is still the problem that these measures are very much tied into the confusing web of popularity, self-confidence, and other indicators of adolescent school performance.
I wonder how popularity correlates with minority ethnic backgrounds and low family income, which are both indicators of lower grades. Kids for whom English is not a first language, for example, tend to get worse grades. But are they less popular? Perhaps in schools where most kids do speak English fluently, but what about schools with very large non-native English speaking populations? And are members of all of these groups less “smart”? I definitely wouldn’t say so. Accepting these indicators of intelligence definitely creates some issues.
Looking at popularity without any reference to race, class, or gender is also problematic. My own experiences indicate that class issues are pretty heavily intertwined with all of this. In my high school, pretty much all of the kids who were noticeably poor were not cool. But there was also some degree of stigma against the kids who were noticeably rich. The popular kids were, almost exclusively and in almost every sense, average.
The problem is that average means different things in different places. Of course we can look at national averages, but my theory is that part of popularity is being fairly average for the particular place where you are. This also raises the question of school size and other skewing factors. Do people in big schools tend to have more friends? There were only 21 kids in my grade in middle school, so I didn’t even have a shot at the 37 required to put me in first place. I wonder how school size correlates with academic performance?
In general, I have a lot of questions about the skewing potential. What would it mean if there is a correlation between the general academic performance of a school and the general level of popularity of it’s students? I could also believe that successful schools tend have less entirely friendless students.
Given all of this, I’m at the point that I often reach with the social sciences: Is this really going to get us anywhere? I am skeptical of any conclusion that we reach being anything more than a nominally educated guess. But, since I have come this far, I’ll try to continue to lay out definitions from which we could move forward.
Even if we accept it’s conceptual validity, popularity remains difficult to measure. In Alex’s mini-study, he goes strictly on numbers and determines that there is a positive correlation between number of friends and grades and test scores. It seems to me, however, that popularity is less how many friends you have and more who they are and what they do. I view “popular kids” as a specific social group, just like jocks or nerds or dorks or hicks or anything else. Popular kids were the kids who cared the most about being popular, and found success in meaningless popularity contests, like the “most popular” class superlative, homecoming queens, student council prom court, etc. At the same time, popular kids are also bonded by the exclusion of others. They affirm themselves and deal with their lack of self-confidence by putting down others. Popularity is not just picking up social cues, but using them to administer a beat down on all of those less able to grasp them.
This, however, is hardly unique to popular kids. It is the same as jocks looking down on everyone who sucks at sports, and kids who do well in school looking down on those who do poorly, and kids who can write computer programs looking down on those who struggle to open word processor documents. I believe that each group is formed and defined in large part by some sense of superiority over others. Again, this may shift over time. In early middle school it may be that the popular kids simply begin excluding those that are not popular. But somewhere along the way, it begins to be a more of a reciprocal relationship. Certainly not all groups of equal strength, and often nerdy groups are really excited if someone “popular” wants to hang out or join them in a way that “popular” groups would not treat a “nerd.”
If you accept that popularity is a group and not a number, than Alex’s findings make more sense. Within each group, the kids that do the best in school also tend to have the most friends. It seems likely that these kids also tend to be more outgoing, well-adjusted, self-confidence, and come from a more stable family situation. But this doesn’t mean that they’re smartest, or that they’re in with the in crowd. It just means that they’re pretty on top of things. At the same time, kids who have the least friends are by no means the nerdiest. They probably have much more going wrong in their lives than those kids getting the best grades, or are the victims of prejudice and policy that hurt both their grades and their social life. Difficult family situations definitely are problems both with school and with friends. It makes a lot of sense.
There are still a lot of questions. All of my definitions have the problem of tending to make things seem binary, when in fact these issues of much less black and white. I maintain my suspicion that popular kids are average, with a tendency against those who are extremely smart or dumb, as well as other prejudices. And I’ve gotten too distracted by all of the details of this to have other big opinions right now.
# posted by : 4:27 AM
: 4:27 AM << Home
