Thursday, January 15, 2009

Bad Science in Education

Deborah Meier from Bridging Differences has a good post about painting a picture of what education could be, as a means of motivating society to put forth the necessary resources to attain it.  She references Charles Murray's The Bell Curve, a book claiming that differences in education achievement are function of innate intelligence, or IQ, and that white kids achieve more than black kids because they are innately smarter.
We both know that on the biggest question—of human potential—Murray is dead wrong. It takes only one example to prove that point. It is no longer a matter of hope or faith for me, but experience. Although one example doesn’t demonstrate how it can be done on a larger scale.
But this post - by a pretty big figure in the education community - makes an elementary statistical error.  Murray claims that the white kid bell curve is shifted several points to the right of the black kid bell curve, so that the typical white kid is smarter than the typical black kid; he does not say all white kids are smarter than all black kids.  In fact, the graph below demonstrates that his own argument requires that almost 50% of black kids are smarter than almost 50% of white kids.

Now, I happen to think Murray is wrong too.  But when people make mistakes like Meier, it makes people like Murray appear more credible, and I think all of this underscores the lack of attention the education community places on mathematical or scientific literacy (perhaps because so few of us have math and science backgrounds).

The more compelling argument against Murray is one mentioned by cognitive scientist Stephen Pinker in The Blank Slate: which is that while population sub-group IQs can be different at a given point in time, they tend to converge in the long run.  And in fact, this is what has happened to most American immigrant groups, suggesting that differences in sub-group IQs are environmental rather than innate.

Wednesday, January 14, 2009

Is Maureen Dowd Necessary?

In today's column for the New York Times, Maureen Dowd remarks that "Hillary aced her Senate hearing," adding:
She was on top of all the issues, no matter how obscure. She batted around our “stale” arctic policy — who knew? — with Alaska’s Lisa Murkowski, who doesn’t seem to realize we’re sick of Alaska.
"Obscure?" "Who knew?" Really?

Russia's land grab in the Arctic was on the front page of the paper Dowd writes for and also made the cover of the Economist. A search for "russia arctic flag seabed 2008" on the New York Times website yields 156 results. "Russia arctic flag" on Google gets more results than "Maureen Dowd." The event has important implications for both our relationship with Russia and the future of energy security. You know, the future of American foreign policy and the defining issue of the domestic agenda.

Thank you, "Newspaper of Record," for this twice-weekly dosage of ignorant sass.

Edit: Someone suggested that Dowd was being facetious. This is unlikely, because she also writes, "[Clinton] was up to date on the inevitable Law of the Sea Treaty," poking fun at the apparent obscurity of the law.

Monday, January 12, 2009

Klein and Sharpton Write Obama

In today's WSJ Joel Klein and Al Sharpton give the President-Elect two education policy recommendations: develop national standards and devote most of the federal money we have now to recruiting and retaining high quality teachers in high-need areas.

First, the federal government, working with the governors, should develop national standards and assessments for student achievement. Our current state-by-state approach has spawned a race to the bottom, with many states dumbing down standards to make it easier for students to pass achievement tests. Even when students manage to graduate from today's inner-city high schools, they all too frequently are still wholly unprepared for college or gainful employment.

Second, the federal government should take most of the more than $30 billion it now spends on K-12 education and reposition the funding to support the recruitment and retention of the best teachers in underserved urban schools. High-poverty urban schools have many teachers who make heroic efforts to educate their students. But there is no reward for excellence in inner-city schools when an outstanding science teacher earns the same salary as a mediocre phys-ed instructor.

Sunday, January 11, 2009

Keeping Google Honest

Google's Public Policy Blog has a new story about how its maps feature Street View helped police locate a kidnapped child.  I don't know if you'll hear them reporting on how it helps criminals scan city apartments for units without windowbars.

Neat Anti-Testing Summary

Quoted from Schools Matter:
As Cal State professor, Art Costa, has said, "What was once educationally significant, but difficult to measure, has been replaced by what is insignificant and easy to measure. So now we test how well we have taught what we do not value."

FT on Google Grand Strategy

One of my inaugural posts on this blog - Google Docs and the Future of Computing - was about how I ditched Microsoft Office in favor of Google Docs, and how Google is shifting computing from users' PCs to company servers. The Financial Times summarizes exactly how this plays into Google's larger mission:

Gmail, its web-based e-mail product, has become a development focus for the Silicon Valley company, with a stream of innovations leading to its promotion by influential early adopters.

With links to Google Docs, Calendar and other web-based services, Google appears to be making Gmail the centre of an online productivity suite that could eventually challenge the dominance of Microsoft’s Office collection of programs.

Google has two motives for pushing harder into e-mail. First, Google makes money from advertising placed inside e-mails. But the service is also valuable for its “stickiness” in increasing users’ dependency and time spent on other Google products.

Microsoft has its own array of Windows Live and .Net products to rival Google, but the integration of Google's free library of applications to the central Gmail platform is what sets it apart.

Comment on Scarsdale and AP

I got a new comment on Scarsdale's decision to ditch the Advanced Placement curriculum that I want to respond to. Here's the comment:
No offense, but this post misses some important points. Scarsdale is one of the most well off school districts in the country.
I agree, at a very high level of academic achievement, standardized testing winds up inhibiting the very best teachers and students.
But most schools aren't Scarsdale. These schools are not hitting the ceiling that standardized tests create.
The problems with the AP curriculum and other breadth-intensive standardized tests isn't limited to just the best schools. To see why, think about why we even want students to study, say, history. Here's my short and inexhaustive list:
  • It allows them to draw lessons from the past to inform the present
  • It develops their critical thinking, analytical, and writing abilities
  • It helps them become better citizens
  • It develops a foundation of historical and cultural literacy that enhances and informs future learning
Standardized tests and the curricula they impose are inconsistent with these ends. These exams cover an incredible amount of information and offer little opportunity to "do history" in depth. This in depth work is what meets those four objectives.

The problem with standards requiring mastery of hoards of facts is that students spend too much time on remembering and understanding information, and too little time analyzing, evaluating, and creating it. This comic from xkcd illustrates my point:

Most people learn the most and learn most meaningfully when engaged in deep and interesting material, and when they learn by doing I learned more about politics and government from high school debate and mock trial than I did in US History or Government; I learned more about statistical analysis doing my undergraduate research than I did in econometrics or my political science research course.

History as an academic subject is valuable largely because of its applicability to the present. Robert Samuelson reminds us that the seemingly obscure theory of mercantilism informs current international trade behavior; a recently declassified history of the use of nuclear weapons informs our analysis of nuclear deterrence. And Thucydides' history of the Melian dialogue is still probably the best explanation of how states behave in the international system.

Learning history is not nearly as exciting or applicable when kids spend most of their time "remembering" and "understanding" facts. Granted, higher level analysis requres a foundation of knowledge, but the point is that current curricula have these out of proportion.

Many kids come in to high school with literacy skills too low to both (a) remember enough facts to pass the test and (b) engage in more meaningful learning. Student creativity, student led learning, and higher order thinking become the province of rich suburban kids. I think it's a worthwhile trade off to require a narrower breadth of knowledge for the sake of deeper learning.

But suppose that all kids came into high school with Scarsdale-like literacy levels. I know that at least for me, I would have preferred history classes that were student-driven and project-oriented. I would gladly give up 50% of the AP Euro fact-regurgitation requirement for projects allowing me to explore cool parts of history. Its easy to make the argument that this is what electives are for; but if you visit at an at-need school, you'll see that its hard to do this when there's so much institutional pressure to focus on the test.
When opponents of standardized tests come up with a better way to accurately and consistently assess student performance across schools, I'll gladly embrace it. Til then, I'll continue to support AP, SAT, ACT, NCLB, etc. - Dank
I agree we need easily comparable measures of student performance for accountability and resource sharing. I like how the AP Art exam requires a portfolio of student work; I think something comparable should exist for other classes. Standardized assessments can be modified to accommodate the outputs of constructivist learning in an objective fashion. As an idea of what this would look like, think about how theses and creating writing classes are graded. At my school, student project grades are determined by an average of many different teachers' rubric scores.

Saturday, January 10, 2009

Glendale Restaurant Serves Heart Attacks

The Heart Attack Grill serves 8,000 calorie "quadruple bypass" burgers (4 half-pound patties) with fries cooked in lard. Scantily clad nurses monitor your vitals while you eat. Thanks to Dan for sharing this.

Watch CBS Videos Online

Friday, January 9, 2009

No Child Left Behind

A recent article in NEA Today highlights some important trends in national test scores and argues that NCLB hasn't made anything better. Take a look at these 2 graphs:

Some thoughts on how this happened:
What happened to make that possible? The War on Poverty? School desegregation? Shrinking class size? Head Start and other new school programs? Maybe all of the above.
Now compare that to education reform in the era of NCLB:

Thursday, January 8, 2009

Live Blogging the BCS National Championship

One of the things I love about college football is the variance in team quality and athletic talent.  In the NFL on the other hand, teams are pretty equal.  After watching a good Mizzou team play all season, it's fascinating to watch just how fast this Florida-Oklahoma game is moving.  The disparities in athleticism really make great contests stand out.

On Getting a New Laptop

A buddy just asked me for some input on getting a new computer.  I thought I'd share my 2 cents publicly as well, so here is my response.

the [linux] font issue has been resolved.  if you get a computer with the intent of running linux you should make sure all of the hardware is linux compatible.  between pc and mac, if money weren't an issue, i would easily get a mac.  their hardware is way more dependable, and the software is designed with the hardware in mind; microsoft has to make windows run on every hardware configuration imaginable, and that is the root of all its evils.  mac os is also better.

i definitely would not get a dell.  the canadian government issued a class action lawsuit against them for systemic overheating across a wide range of models; i was part of and won a separate class action against them when i was in college.  my current dell is also having problems.  i am measuring the computer temperature at 62c idle when it should be 30 or 40.  the computer dell gave me to replace my old one with the class action lawsuit was bequeathed to my parents, and it overheated just before christmas.  if you google "dell hell" you will hear many more stories and lawsuits.

dell offers great bang for the buck, but it really is a gamble.  and after having 2, i don't think the cost effectiveness is worth the potential headache.  if i were to get a pc i would probably get a lenovo thinkpad, the enduring pc legacy of IBM.

Letter to a Marxist

I recently had a discussion about systemic explanations for education inequality with someone who said that a main (or the main) reason why we have systemic education inequality is that the ruling class does not want an educated proletariat for fear of it disrupting the social order. In other words, if poor people were educated, they would uproot the social structure, make it more equal, and cut into the wealth of elites. We both stipulated that America has the resources to radically improve education, but isn't doing it. This is part of an email I wrote in response:
My problem with this claim is that I think it doesn't have the same evidential support of more simple, competing theories: namely, subtle racism, the self-interest of taxpayers, and structural policy failure.

At the most simple level, we can observe visible incidences of racism: racial epithets, lynchings, hate crimes; and we reasonably infer that less visible racism exists. I grew up in very blue collar communities, and I got the general sense that a lot of blue collar white people feel they get taxed for working hard so minorities can get handouts for being lazy. There is also a solid body of economic research explaining wealth redistribution vis-a-vis welfare on racial terms: people of race A are more likely to support wealth redistribution if the recipients are also people of race A. I also think you underestimate most Americans' dislike of the taxes requried to improe education. Finally, I think think there are also systemic or structural forces that undermine good education policy. For one, the money needed for the reforms you cited is only available at the national level, but the structure of education is that its locally controlled. Many people oppose uprooting local control on the grounds that locally controlled schools allow for flexibility, or on federalism grounds. Up until recently, any such federal intervention would have been vigorously opposed on federalism grounds to protect federalism so states could be racist. And many, many conservatives in congress philosophically oppose the broad reach of the federal government (it's why most of the legal support for the medical marijuana case Gonzales v. Raich came from the conservative legal community on 10th Amendment and interstate commerce clause grounds).

Having said all this, I have never heard anyone "corporate" explicitly say or otherwise indicate that they don't want poor kids to be educated; but I have heard people say "black people are lazy,"; and Rush Limbaugh has publicly broadcast that America's drug problem is the product of weak-willed minorities. On the other side, many corporations and figureheads give back to education. The President of Goldman Sachs (Jon Winkelried), for example, is personally very involved with education because his son is special-ed. Small schools - for better or worse - are partially funded by the Gates Foundation. And corporations systemically donate to education over many other philanthropic causes.

Now, it's possible that these are all features of the superstructure and that Winkelried and Gates have deluded themselves with a false consciousness, but I can't help but be sceptical of such a non-falsifiable proposition (however meaningful the narrative or elegant the theory) in the face of simpler, empirically supported theories.

I guess my question to you is, what reasons do you have for believing that this Marxist oppression hypothesis is a better explanation of education inequality than these other simple - and it seems to me better evidenced - explanations?

Teach for America and Teacher Quality

Over the last 3 years my school hired 6 Teach for America teachers and 2 left before completing their two year commitment (one after a year, one after a semester).  Many education reformers believe the quality of education is inadequate because the best and brightest choose other professions over education and so the remaining pool of teachers is less than what we need.  TFA recruits and accepts people largely based on GPA and extracurricular achievement; the best and the brightest will be the best teachers.  And it's a compelling argument that an ivy graduate will learn a new skill faster than someone from a community college.

The two teachers who quit were truly cream of the crop: great schools, great achievements, great resumes.  (I don't know if other teachers quit at lower or higher rates.)  TFA in NYC averages about a 90% retention rate over the 2 year committment.  While external achievement may proxy successfuly for teaching achievement in the aggregate, there is a lot of variation too.

Many educators say that it takes 5 years to be a good teacher, and despite all of the public rhetoric on intrinsic ability determining teaching quality, both the Obama and McCain education plans focused on speeding teachers' ascent of the learning curve. 

Here is the text from the Obama plan:
  • Expand service scholarships to underwrite high-quality preparation for teachers.
  • Support ongoing improvements in teacher education to enable teachers to meet the challenges of their demanding jobs,
  • Provide mentoring for beginning teachers so that more of them stay in teaching and develop sophisticated skills.
  • Create incentives for shared planning and learning time for teachers.
  • Support career pathways in participating districts that provide ongoing professional development and reward accomplished teachers for their expertise.
And here is the text from the McCain Plan (go figure the McCain website is down, but I still had his education platform saved as a pdf):
Provide Funding For Needed Professional Teacher Development. Where federal funds are involved, teacher development money should be used to enhance the ability of teachers to perform in today's technology driven environment. We need to provide teachers with high quality professional development opportunities with a primary focus on instructional strategies that address the academic needs of their students. The first 35 percent of Title II funding would be directed to the school level so principals and teachers could focus these resources on the specific needs of their schools.
One of the things that makes me uncomfortable in the education community is that these two approaches to maximizing teacher quality are often seen as mutually exclussive.  I hardly ever hear people in the TFA camp acknowledging the importance of experience and professional development, or education traditionalists even acknowledging the existence of intrinsic human capital (excpet indirectly when complaining about other teachers).

Even supposing an extreme version of the alternative-certification narrative (read: teaching achievement is primarily determined by intrinsic ability as measured by traditional indicators), there is still an upper limit on the amount of the "best and brightest" we can lure away from other fields.  To wit: even if we get all the best and brightest, they will still be only a fraction of the overall community of teachers, and the traditional focus on teacher quality improvement, classroom-size reduction, etc. still seems necessary to close the achievement gap.

Groundbreaking Charter School Research

There isn't a lot of rigorous scientific research in education.  A big problem is an inability to conduct truly randomized studies; another problem is that ed schools don't really train teachers in scientific research and statistical analysis.  So when stuff like this comes around it deserves our attention.  A Harvard Education Professor teamed up with an MIT Economics professor to come up with one of the better studies I've seen.

Key findings below:

For elementary Pilot School students, a significant impact was seen in English Language Arts scores, but not for math scores.  In middle school, the observational results suggest Pilot School students may actually lose ground when compared to their peers in traditional schools, while the lottery-based results showed no difference between Pilot School and traditional school performance. At the high school level, observational results showed significant improvement of performance by both charter and pilot school students, compared to student performance in traditional schools. The lottery-based study, however, showed no significant difference between high school students in Pilot School and high school students in traditional schools.

Among other key findings of the report: the impact of charter schools was particularly dramatic in middle school math. The effect of a single year spent in a charter school was equivalent to half of the black-white achievement gap. Performance in English Language Arts also significantly increased for charter middle school students, though less dramatically. Charter students also showed stronger performance scores in high school, in English Language Arts, math, writing topic development, and writing composition.  Students in pilot high schools also made measurable progress.

Wednesday, January 7, 2009

Terrorist Attacks on Gay People?

The Seattle Times reports that anonymous letters threatened gay bars with ricin attacks:

Eleven gay bars in Seattle were sent letters Tuesday threatening ricin attacks — in what some are describing as a hate crime.

The anonymous letters say, "I have in my possession approximately 67 grams of ricin with which I will indiscriminately target at least five of your clients. ... I expect them to die painfully while in hospital."

A 12th letter was sent to the alternative weekly The Stranger, according to its Web site. That letter says the paper should be "prepared to announce the deaths of approximately 55 individuals."

It's worth noting that in the documentary Jesus Camp, children's pastor Becky Fischer explains that little boys and girls should show the same faith and commitment to Jesus as children who strap bombs on their chests for Islam.

Recession Hits Detroit Elementary School

Maybe education needs a bailout:
A Detroit elementary school is asking for donations of toilet paper and light bulbs to keep their school functioning.

The principal of the Academy of Americas sent a letter to staff, parents and partners asking for donations of items "that are of the utmost importance for proper school functioning and most importantly for student health and safety."

In the letter, Principal Naomi Khalil cited budget constraints within the district as the reason why the school could no longer stock the items.

Switch to Linux

I decided to take the leap from Windows to Linux.  I installed Ubuntu (free of charge) on one laptop (that I use like a desktop) but held back on my smaller (more portable) laptop for fear of not being able to run Stata.  But yesterday I managed to install Stata 10, so today I'm formatting my second computer to linux as well.

My reasons for the switch:
  1. Google Documents has removed my need for Microsoft Office
  2. Most applications I use I run through a web-browser (email, calendar, office software, grading, etc.)
  3. No crashes, viruses, malware, spyware, or adware
  4. Workspaces allow me to run multiple "desktops" on one computer.  I have one desktop for Google Applications, one desktop for work, one desktop for play, and I can create more for any of a number of specific tasks.  I don't need any new program to do this; it is all seamlessly integrated and pre-packaged with the Ubuntu shell. 
Here is an image of switching workspaces:

You can set your screen to show the workspaces as faces of a cube, cells on a filmstrip, or tiles in a window. 

My one reservation so far is that fonts don't always display as neatly in web pages as they do in Windows, and sometimes they look downright unattractive.