How there can be and NOT be an educational crisis in the US

September 25, 2010 at 1:17 pm 8 comments

Back in the late 1980’s, there was a decline in the SAT scores in the United States.  Critics moaned about the woeful state of our educational system.  A really interesting report came out from Los Alamos showing that, for all socioeconomic groups, SAT scores were actually rising.  What happened in the 1980’s was a dramatic increase in the number of poorer (low socioeconomic status) groups taking the SAT, and those new kids started out with lower scores (but improved every year).  That large influx at the low-end drove the average down, even though everyone was learning.  The average alone didn’t tell the story. The high variance should have alerted the critics to a more interesting story in the distribution.

That’s what I think Nicholas Lemann is pointing toward in his article.  The counter-arguments to Lemann’s case, expressed in the comments, were well-founded (except for the ad hominem attacks), but do not actually refute what he’s saying.  There are several reasonable scenarios where both Lemann and those concerned about the crisis in American education are right.

Let’s imagine a picture of the distribution of knowledge 100 years ago in the United States.  Here’s my back-of-the-envelope (actually, finger-sketched on the iPad) graph of what the graph might look like, with knowledge increasing from left-to-right and number of people being educated in that knowledge graphed on the vertical axis.

The people who are educated are really well-educated, shifted to the right of the graph.  But the graph drops off rapidly.  There were lots of people in the United States 100 years ago, but very few of them would appear on the graph.  Most citizens in the United States were not educated at all.

There are lots of possible graphs for today, but here’s one that I think matches a lot of the available data.

There are people in the United States today who are educated better than the best were 100 years ago, but few of them.  Our rightmost peak is smaller than it was 100 years ago.  But now we have this big hump in the middle.  The vast majority of people are getting some education today.  I suspect that there is another hump to the left in our graph, the folks on the other side of the Digital Divide, the ones who have been most hurt in the current recession.

Given these two possible curves, one can understand the “There is no crisis argument.”  We have a lot more people under the curve today, than 100 years ago.  The vast majority of people in the United States are literate (to some level) and have received some formal education.  In terms of creating a better workforce, a more educated quantity of people, that’s an improvement.  I suspect that our area under the curve is greater in the United States than in much of the developed world because of the large numbers of educational opportunities in the United States, but I don’t have evidence for that.

It’s also possible to see multiple crises in the second graph.

  • The average in the second graph is lower than in the first graph.  Sure, but I’m less concerned about that. It’s shifted the same way that distance education courses get students to do as well as in-class students: Get rid of all those on the left of the distribution.  Now, the average may be lower than our society needs in today’s graph, and that would be a reason to be concerned.
  • If we want more innovators, if we think we need more people who will push the envelope, we’re in trouble.  We have fewer people than we need on the right side of the graph.
  • If that big hump is too far to the left, “below threshold” as Alan would say, then we’re in trouble.  Our knowledge-driven workplace may need that majority-knowledge level to be greater.
  • If there is a second hump to the left, and if there are lots of people in the United States who are not under the curve at all, that’s social, perhaps criminal negligence.  Even if it’s not in the Constitution, as Lemann points out, public education is a right in the United States.

For those wondering, “When is Guzdial going to get back to talking about computing?!?” — I think I am.  I believe that the 100 year ago curve represents pretty well the current state of computing education.  We have a peak of people who know something, but it drops off quickly, and most people aren’t under the curve at all.  I would like to see us looking more like the bottom graph.  There is a need for the vast majority of educated people to know something significant about computation.  We need more people under the computing education curve, and I’d even take that middle hump.

Entry filed under: Uncategorized. Tags: , .

New NSF Program: Computing Education for the 21st Century WGBH, ACM Celebrate Dot Diva Launch to Reshape Image of Computing for High School Girls

8 Comments Add your own

  • 1. Tom Hoffman  |  September 25, 2010 at 1:37 pm

    Not that it matters to your point about computer science, which seems pretty plausible, but your sense of educational attainment in America in 1910 vis a vis 2010 is completely whacked. It isn’t like Edison had a Ph.D. from Carnegie Mellon.

    • 2. Mark Guzdial  |  September 25, 2010 at 1:42 pm

      If you’re arguing that higher education isn’t necessary to be an innovator, I agree. My argument is about numbers of people and knowledge, not individuals and not degrees.

  • 3. David Klappholz  |  September 25, 2010 at 3:46 pm

    I don’t know if there’s a general crisis in education or not — except in very poor communities; but the reason for the timing of the report that started this discussion is the current economic crisis, especially the high unemployment rate among non-college-educated people, not any convincing educational statistics.

    Given what I’ve read of the report so far, i.e., just its summary, I’m not sure if we’re in trouble because we’re doing less well educationally than we have been or because China is doing very much better than it has done until recently…and because every time someone in the US comes up with an innovative product — and the pace at which this is happening hasn’t slowed down — we contract with a factory in China to do the manufacturing. This loses us both manufacturing jobs and all the secondary and tertiary jobs associated with manufacturing. (The iPad on which Mark drew the graphs is a great example. Apple employees invent a great new product, and they’re the only Americans, other than, possibly, shareholders, who benefit.)

    I’m not an economist, so I don’t know how significant a role this plays — and I’m not quite sure that even our best economists do either — but it’s the only factor that looks convincing to me.

  • 4. Katrin Becker  |  September 25, 2010 at 5:17 pm

    You mention that you think there are more people going to school in the US than elsewhere because the US offers more opportunities. The following report gives some comparisons.

    This report includes 17 developed countries and lists many measures such as HS & university completion as well as various skills. Interestingly, the US ranks very high in HS and university (but not college) completion, but when it comes to reading, math, science, and problem solving skills, it ranks poorly. One interpretation of that is that although there are lots of people getting through school, they are not learning as much as people in other countries are.

    That’s where the crisis seems to be – it is not in the number of people getting through school, it is in what they are or are not getting out of it.

    • 5. Mark Guzdial  |  September 26, 2010 at 1:23 pm

      That’s what the middle hump in my “today” graph is meant to represent, Katrin. It may be (as Alan suggests) that we are educating lots of people, but not to a high enough level. Other people may have their “middle/majority hump” shifted to the right, but they may have smaller numbers under the curve over all.

      • 6. Erik Engbrecht  |  September 26, 2010 at 2:23 pm

        Are you sure the problem is that they are not getting enough knowledge, as opposed to something else?

        I’ve noticed a lot of people seem to expect work environments to be as well structured, with as clear of expectations and feedback and as clear of a progression of advancement, as school. I just don’t think that happens in many work places. And if it did, the company would basically be declaring “we don’t want any innovation. period. end of story.”

        All the brilliant, highly knowledgeable people in the world won’t do us any good if they’re a bunch of conformists looking to please authority figures.

  • 7. Alan Kay  |  September 26, 2010 at 8:17 am

    Hi Mark,

    I think you and some of your commenters are touching on the real issues.

    My criticisms of Lemann’s piece have to do with both content and rhetoric.

    I think today there are more opportunities for a wider range and variety of Americans than ever before. This is good, but is not the main or a good criterion for assessing the quality of US education.

    I’ve already commented on his error regarding “a faith in the capabilities of ordinary people that the Founders didn’t have” via a famous quote of Jefferson to the contrary. The Founders did worry about “tyrannies of the majority” and in direct democracy for a wide variety of reasons, but they were firm that the entire voting public should have the “ultimate powers of the society”.

    One of the aims of general schooling was to provide the “wholesome discretion” needed by this voting public. In my view, we do not have this today. This is one of those threshold arguments — that if a system has thresholds then improvements below a threshold don’t matter.

    The other very important larger dimension beyond citizenship is again one of thresholds. Is the current educational system up to today’s needs?

    Lemann in several places tries to somehow equate demand with quality. But this is specious. Demand for something often can lower quality in the effort to make more of something that can be sold to more people — especially if the something is a piece of paper that can be used to more easily find a job.

    It is very hard to both get good enough data from the past and to normalize it well enough to allow real comparisons with the present. But the real issues here are about the poor state of the present and not whether this state is better than the past.

    Best wishes,


  • 8. Alan Kay  |  September 28, 2010 at 1:10 pm

    Hard to leave this one alone, especially given Lemann’s CV.

    There is no question that many more opportunities of all kinds, including to become educated, exist in our society today. But the real questions (to me at least) have to do with thresholds reached or not for important roles in life (for example being a voting citizen in a representative democracy).

    It is extremely difficult to assess levels of education for the general population in the past (see e.g. the classic History of Literacy in the US by Kaestle, for what,why and how).

    But here’s one of my favorites, for which good numbers are known.

    In Jan 1776 Tom Paine published “Common Sense”, a 48 page simply written argument against monarchy as a form of government. Here is what happened as excerpted from the Wikipedia article.

    Thomas Paine began work on Common Sense in late 1775 under the working title of Plain Truth. With the help of Benjamin Rush, who suggested the title Common Sense and helped edit and publish, Paine developed his ideas into a forty-eight page pamphlet. Paine published Common Sense anonymously because of its treasonous content. Printed and sold by R. Bell, Third Street, Philadelphia, it sold as many as 120,000 copies in the first three months, 500,000 in the first year, and went through twenty-five editions in the first year alone.[5][6] Paine donated his royalties from Common Sense to George Washington’s Continental Army, saying:

    As my wish was to serve an oppressed people, and assist in a just and good cause, I conceived that the honor of it would be promoted by my declining to make even the usual profits of an author.[7]
    —Thomas Paine

    Pretty cool isn’t this? Now the big whammy is that there were only about 1.5 million men, woman and children (non-slaves) in the 13 colonies, and the final press run over the first several years is estimated to be about 900.000 copies.

    Did they buy these to put on their coffee tables (as apparently happened with Stephen Hawking’s book?)

    Probably not. It is far more likely that the other high estimates of literacy in the 18th century in the US are accurate, and that a very large percentage of the colonists did read this pamphlet. We do know that the pamphlet was one of the most important unifying forces that preceded and carried along the Declaration of Independence.

    Two sobering thoughts for today.

    1. 25 years ago the National Lteracy Foundation did a study that showed only 20% of American adults today could read “Common Sense” for content as literate readers.

    2. There are 300* Million Americans today, and there is no printing resource that could reach 150-200 million of us today. Only the Internet can do this, but nothing is set up to generally be able to fluently read that level of essay today.

    An important question is “How could Lemann not be aware at least of the history of “Common Sense”, etc and what it signified?

    Best wishes



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Trackback this post  |  Subscribe to the comments via RSS Feed

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 7,142 other followers


Recent Posts

Blog Stats

  • 1,729,803 hits
September 2010
« Aug   Oct »

CS Teaching Tips

%d bloggers like this: