Schools of Thought

EdWahoo is going on haitus until the end of January as I devote my time to the last two weeks of my term and the transition period at The Cavalier Daily. Look for posts to resume Feb. 1.

I've been thinking a lot lately about just how much goes on in a school. Specifically, stemming from conversations with friends and family members who are educators and my own experiences shadowing teachers for a day in a local elementary school, the not novel but often forgotten aspect of non-academics. I'm not simply talking about the social network, but everything that gives a particular school its character -- major policies, how safe the halls are, how much of a community exists, how happy the students are, how happy the teachers are, etc. These are factors that never show through on even the best assessment, whether it be standardized or performance-based.

As Noel Epstein put it:

They not only provide before-school programs, breakfasts, lunches, after-school care, afternoon snacks and sometimes dinners (as well as summertime meals). They also instruct children about sex and, in many places, teach them to drive. They face growing pressure to take tots as early as age 3 in pre-kindergarten programs. They share responsibility for keeping children off drugs, making sure they don't carry weapons, instilling ethical behavior, curbing AIDS and other sexually transmitted diseases, battling alcohol abuse, preventing student suicides, discouraging cigarette smoking, tackling child obesity, heading off gang fights, providing a refuge for homeless children, ensuring that students are vaccinated, boarding some pupils, tending to toddlers of teenage mothers and otherwise acting in loco parentis in ways not anticipated a generation ago.

Though critics bemoan this trend, there's little chance of fundamentally altering it, for several reasons. Chief among them is that schools generally are reacting to what the public wants. Many people seem to think that adults' worries about schools center mainly on student achievement. That's wrong. While test scores certainly keep business, political, media and other elites up nights, they are not what most trouble the wider citizenry, as polls have long shown.

According to a Public Agenda analysis of opinion surveys, for example, Americans in 1999 said that the top three problems facing public schools were lack of parental involvement, drug use and undisciplined students. Academic standards came in seventh. Similarly, that year's annual Gallup education poll found far more concern about violence, gangs and other student behavior than about academics, which trailed in ninth place. By last year, when Gallup ranked the public's top five school concerns, academics were not cited at all (inadequate funding led the list), and this year's poll showed again that student achievement wasn't among the public's main worries.

Now, it's obvious to me that the first and foremost goal of schools should be the academic education of its students, since that's where the equality of opportunity comes from. But there's so much else that goes on in a school which directly affects the academics that it can't be profitably ignored. An educator I know was telling me about her high school, which serves a poor, high-minority population and actually does amazingly well under the circumstances. Still, she said, there are parts of the school which her female students will not enter alone. Similarly, another educator's elementary school recently found gang graffiti in the bathroom which led to a great deal of fear among the faculty and parents and caused new rules such as requiring whole-class bathroom breaks.

These auxillary factors don't often come up when judging a school's effectiveness, but as we move into the next generation of accountability (growth models, etc.) it seems to me that school profiles would provide a rich source of context. Some school report cards in California, for example, do report the number of recorded fights; but of course, the minute you tie consequences to number of recorded fights, the number of recorded fights is going to plummet. Eventually, the ideal would be massive data banks with every concievable metric about a school, thus allowing interventions to be better targetted and subtler problems to be found and addressed. Some of these, no doubt, are going to have to have more subjective pieces, but that's ok. So long as the data isn't high-stakes, there's no danger in subjectivity and there's little incentive for dishonesty.

Test scores are a fine first-look sorting mechanism. But how safe is the school? How involved are the parents? How is the principal judged by students, staff and the community? How much agency do the teachers feel they have? The students? A school is more than lessons and tests, and it would be worthwhile to actively acknowledge that.

I had some illuminating experiences observing at my mother's elementary school Monday, but i'll post about those soon. For now, I want to talk about the column heard 'round the eduworld -- George Will's jeremiad against Education Schools in Newsweek. Will argues that Ed Schools are counterproductive largely because,

The dogma has been that primary and secondary education is about "self-actualization" or "finding one's joy" or "social adjustment" or "multicultural sensitivity" or "minority empowerment." But is never about anything as banal as mere knowledge. It is about "constructing one's own knowledge" and "contextualizing knowledge," but never about knowledge of things like biology or history.

I'm not interested in getting into the specific question of Ed Schools; people far more expert and far smarter than I are the commanding voices in that debate. I want to challenge Will on his straw man syllogism that equates the flaws in Ed Schools with flaws in an imaginary ideology of "progressive education."

First, Will is simply incorrect in suggesting that any variant of progressivism/constructivism is the dominant form of pedagogy in contemporary American education. As I've detailed, the National Assessment of Education Progress, High School Survey of Student Engagement, Trends in International Math and Science Study and National Assessment of Adult Literacy all bear out the widespread prevalance of traditional teaching in our schools. Moreover, almost all 50 state assessments -- see Texas, Virginia, California and Georgia for examples -- ask overwhelmingly multiple-choice questions which require almost nothing but (wait for it) knowledge.

Next, Will, ever the crafty writer (having penned about a hundred columns myself, the techniques stand out), sets "contextualized knowledge" and "constructed knowledge" in opposition to "rigorous" knowledge, grouping them instead with the somewhat less academic goals of tolerance and self-confidence. The only problem is, take away context and take away students' ability to learn for themselves, and what you're left with is facts. Quite simply, he is lionizing lonely nuggets of information that are supposed to somehow constitute the skill sets and thinking ability necessary for all children to achieve their highest potential.

I don't doubt that some people espouse the soft philosophy Will and Heather MacDonald assign to everyone who doesn't sleep with a copy of direct instruction manuals underneath their pillow, but those people are a minority and are embracing a wholly bastardized version of progressivism. Piaget made it abundantly, explicitly clear in his writings that pre-existing knowledge was necessary in order for new knowledge to be constructed. Dewey wrote that "Education must be conceived as a continuing reconstruction of experience" and offered that the classroom should resemble a tiny democracy; he never said that students should be left with no structure and no guidance in the hopes they stumble upon thinking skills.

I could go after Will's other arguments, such as pointing out that there are plenty of public and charter schools which take a more progressive approach and succeed just fine (my ubiqutious love of H-B Woodlawn works well as an example here); or perhaps noting that the decline in adult literacy coincides with the rise of the conservative-supported "Standards Movement" (not to make the case that the two are related, only that blaming a certain philosophy while the country was moving in the other direction doesn't make a whole lot of sense), but let's leave it at this:

It is fair to make the proposition that Ed Schools aren't up to snuff. It is manipulative to connect flailing Ed Schools to the unique liberal philosophies many of them espouse, generalize those philosophies, and then declare that this amorphous progressivism is dominant in our schools and responsible for the majority of the problems we're facing. It's not, and I have a hunch George Will knows that.

I loathe multiple choice standardized tests. I took the Praxis II: Social Studies Content Knowledge exam yesterday, and I have to say, my score on that test has only the most minimal of anything to do with a) my content knowledge of social studies or b) my ability to teach it.

I say this because the bulk of the questions fell into one of three main categories:

The absurdly easy (real example: "In which region of the world did the religion of Islam arise?")
The absurdly irrelevant (real example: "What is the longest unfortified border in the world?")
The absurdly arcane (real example: "What is the significance of [some number] latitude?")

I don't know precisely when the idea arose that facts constituted desired knowledge, but it's not helping anyone.

In other Edwahoo news, I've been offered a Teach for America position teaching elementary school in Phoenix, AZ, which I do believe I'm going to accept.

I was quite pleased to see this announcement coming from the Bush administration yesterday:

The Bush administration on Thursday proposed spending $114 million on educational programs to expand the teaching of Arabic, Chinese, Farsi and other languages typically not taught in public schools.

Speaking to more than 100 college and university presidents attending a two-day conference at the State Department, President Bush said the effort would play a critical role in national security and lead to American students' gaining a better understanding of foreign cultures.

[...]

[Asst. secretary of state Barry Lowenkron] said that only 44 percent of American high school students were studying any foreign language and that 70 percent of those were learning Spanish. Ms. Powell said that by comparison, the nation had only 2,000 Chinese language grade-school teachers.

Margaret Spellings, the education secretary, said in an interview that efforts to teach such languages as Chinese and Arabic to children as young as 5 were brand new. "We don't know how to do it. This whole notion is in its infancy. But our hope is this is a start, and we can build on it."

I myself took three years of high school Spanish, and I still think Spanish is an exceedingly useful language given the percent of the world that speaks it, but they're spot on with regards to the dearth of Chinese and Arabic speakers. It's a bit archaic that we still funnell most non-Spanish students into French and German. Don't get me wrong, they're still useful languages, but this is one of those situations where with limited resources and the same basic product you should be tailoring the program to your needs.

I've also always been bemused by the fact that we start teaching foreign languages in school right after the cognitive developmental stage at which children are most ready to learn a new language. It seems to me that, given both the world's increasing globalization and the linguistic cognitive benefits of language, it should be a subject that we start teaching in the earliest grades. I'm just not entirely sure where in the school day to squeeze it in. Though I do wonder about Secretary Spellings' last remark; I'm fairly certain that foreign languages are taught from very early on in other countries -- couldn't we find our models there?

Either way, this is a laudable beginning.

Samuel Freedman has an interesting profile in today's New York Times of liberal activist and staunch pro-No Child Left Behind advocate William Taylor. While some of the heaviest criticism of NCLB has come from liberal quarters, Taylor is not alone in bridging social conciousness and support for the Act; most of my friends at the Education Trust fall in the same category. Let's not forget, also, that NCLB was the culmination of a long legislative proccess, much of which occured under President Clinton.

It's an interesting vice that liberals have been put in. On the one hand, I think many people understand the flaws with NCLB's emphasis on standardized testing and by extension fact-based learning, the myriad ways to circumvent its intentions and the borderline unfunded mandate it at times represents. On the other, it's hard to deny that across-the-board accountability prevents districts from ignoring poor or minority students, thereby opening a road towards equity.

The determining factor for where a liberal falls in that spectrum (which in turn probably dicates one's support or detraction from NCLB) is the question of how much someone thinks schools can do. If you're more the demographic determinist, society-first Richard Rothstein type, then schools can't do very much at all and the benefits of accountability are outweighed by the costs of the imposition. If you think schools can largely make up the educational difference in spite of the community, then crappy standardized tests and some stifling of teacher creativity are bitter pills worth swallowing.

Obviously, these positions are not concrete; the strongest anti-NCLB folks can agree that accountability should be improved while the staunchest NCLB supporter wants to see the community improved. I fall somewhere on a different axis, taking the position that schools can make the difference and that NCLB was/is a necessary short-term program for the purposes of establishing the infrastructure for accountability (data systems, ceasing the rug-sweeping, etc.) but needs to start giving way to pedagogy and related structures designed to impart the right type of skills -- namely critical thinking.

What's really interesting to me is that it seems many liberal supporters of NCLB have fallen back on the argument "there's nothing better," grounded in the aforementioned ideals of equity. Take Taylor, for example:

Mr. Taylor had heard [all the criticisms of NCLB] before, and something combative crept into his tone. "I really have not heard an alternative," he put it. "To say more of the same, just give more money, is not an answer. It is not good enough to say this is a societal problem, though certainly that is the case. If you say everybody is responsible, then nobody is responsible."

A moment later, he threw out a dare: "If you've got some better ideas, then let's hear them. It's time for them now."

First off, when your defense for your policy can be summed up as "it's less bad than everything else," that's not exactly inspiring. But more importantly, Taylor is presenting a false choice. The implication is that you can either have NCLB or a system with no accountability in which poor and minority kids again recieve different and worse treatment than their counterparts. I think you can maintain accountability while alleviating the major flaws of NCLB. In fact, I'm planning on staking a whole lot of time and energy on that proposition. It seems like creating a well-run, well-funded system that maximizes thinking skills while maintaining equity and accountability should be the next educational endeavor, and one all liberals (and everyone else!) should be able to get behind.

Debate is healthy, which is why I've calmed down my visceral reaction to this column by Dan Seligman in Forbes asserting quite firmly that the achievement gap cannot be closed and, by implication, we shouldn't even be trying.

Instead, let's eviscerate it logically.

The equity theme here has two components. One is that the prime objective of educational policy is to eliminate the "achievement gap"--the gap between what's learned in school by disadvantaged kids and what's learned by middle- and upper-class kids. The other element is the notion that the U.S. would be much better off if only we devoted more resources to the education sector.

Pursuing the latter, Henry M. Levin, a professor of economics and education at Teachers College, argued in an op-ed piece in the New York Times appearing shortly after the symposium that immense gains are to be had from keeping underachievers in school. High school dropouts, he noted, earn $260,000 less over the course of their lives than do high school graduates. But it is wildly unrealistic to assume that the students who drop out would earn as much as those who graduate if only they had hung around for four years--an assumption ignoring tons of evidence that the dropouts are a population with trouble hanging on to jobs and have lower ability levels. The data tell us that the average dropout has an IQ that puts him at around the fifteenth percentile of those who graduate. Not exactly what employers are looking for.

Consider that proposition. If indeed the average dropout has a massively lower IQ than the average graduate, and since we know that a wildly disproportionate number of dropouts are minorities and poor, this leads to the simple syllogism that poor people, hispanic people and black people are born stupid and are going to by and large stay stupid.

Of course, proponents of this line of thinking wont put it in those terms. They'll argue that increased levels of lead in the water, more malnutrition, lower parental education, less vocabulary spoken at home, etc., are all the unfortunate factors of poverty which lead to the lower IQ. And, I will readily concede that many poor and minority students enter kindergarden already grade levels behind their wealthier peers.

But here's the thing: Schools can bridge those differences within a matter of years. Studies have shown, for example, that students who have high-quality teachers three years in a row can rise multiple standard deviations in achievement. Or, you can look at individual schools that serve high-poverty, high-minority populations yet remarkably get achievement rates in the 90%+ region: Dayton's Bluff, M. Hall Stanton, Elmont, City Heights... just in Virginia, nine elementary schools had at least 50% minorities and 75% poor students yet achieved in the top 25% of the state in 5th grade reading. That may not seem like a lot, but 500 elementary schools across the nation tell you in the most concrete terms that poverty and skin color are not shackles attached at birth.

Also, I don't know, perhaps the persistent achievement gap has something to do with the fact that disadvantaged schools get on average worse teachers with less experience, lower expectations and fewer majors in their subjects; buildings which are often crumbling and don't contain adaquate technology or resources such as books; and, on top of all that, recieve on average $1,300 less per-pupil.

After all of that, we can start arguing about the flawed use of IQ as an absolute immutable measure of intelligence, whether with Howard Gardner's contentions regarding multiple intelligences or the many studies that have shown inherent biases against minorities in IQ tets.

Of course poverty brings with it a whole host of challenges. But to suggest that those challenges are insurmountable -- and then, what, we give up? tell the hundreds of thousands of poor children that they better start preparing for a life flipping burgers (if they're lucky) because we're not willing to put in the resources that we know can help them reach an equality of opportunity? That the American Dream has a "poor need not apply" sign hanging outside it's window? -- is dead wrong.

What's so strange about all this? Just one little thing: It is not possible to close the achievement gap. The mission statement is a summons to a fool's errand. The reason that the gap will never be eliminated is that intelligence rises with socioeconomic status. Estimated correlations between social class and IQ range from 0.3 to 0.7 (on a scale where 0 means no connection and 1 describes two variables marching in lockstep). Those figures tell us that the poor and disadvantaged have less cognitive ability than those from higher-status families. Cognitive ability predicts scores on achievement tests.

Note that this is Seligman's sole evidence for his sweeping assertion that the gap is a permanent inevitability. Should we point him to Delaware, where the black-white achievement gap has closed 7 percentage points in the past few years? How about Pennsylvania, where it contracted by 8. In Kentucky, the income achievement gap went down 5% between 2002 and 2004. Same in North Carolina. Even the long-term NAEP results bear out that the achievement gap lines are dynamic. If Seligman was right, we should expect little to no movement. I've long been a critic of the current system and I've long said those numbers aren't shifting with nearly enough vigor, but the very fact that they are shifting under the weight of efforts to shift them is powerful testimony against the defeatist equation of poor = dumb.

Again, I also have to come back to what education should be teaching. Seligman offers that "Everyone hits a brick wall at some point. With some students it may not happen until they are exposed to quantum mechanics. With others it happens with long division. Most students are well inside those two extremes, but the fact remains that disadvantaged students hit the wall earlier and learn less." I'm not quite clear what the purpose of schools are if not to provide a helping hand over that wall. Moreover, if instead of trying to teach everyone quantum mechanics we were trying to teach everyone to be efficacious critical thinkers with the skills necessary to have opportunities, well, maybe we could lay a path that would avoid the wall altogether.

I come back to this: A child born into poverty has a 9% chance of recieving a bachelor's degree by age 25. Either that means that the bulk of poor children are morons, the bulk of poor children are screwed, or there's something wrong with the schools.

Considering that I can drive a half hour from my house and go into a school that's graduating 90%, not 9%, of it's poor students, I'm going to keep fighting to give every student a shot at achieving their highest potential.

[More:
Eduwonk here
EdTrust's Weiner on teacher quality here
Cleveland H.S. students on the challenges to education they face in school here]

I've long made the contention that most multiple-choice standardized tests don't assess high-level critical thinking skills such as evaluation, judgment, inference and connection of disparate information. At the same time, I've also held that widespread assessment is crucial and necessary for an accountability system which ensures all students are being well served. A commenter below asked what my ideal standardized test would look like. By way of beginning to answer that, I think it's worthwhile to examine better and worse types of exams.

I loathe multiple choice items, and there's a reason -- the vast majority of them test nothing but basic information recall or one-step understanding. For example, these question from the 2004 8th grade Virginia science SOL:

15. Which of these planets in the solar system was the most recently discovered?
A Mars
B Venus
C Jupiter
D Pluto

20. Which biome contains large populations of grazing herbivores, few species of birds, and deep, rich soil?
F A taiga
G A tundra
H A deciduous forest
J A grassland

We can cross the country to see this at work in Oklahoma's 8th grade history end-of-course exam:

19. What battle convinced British leaders to begin the peace talks that ended the American Revolution?
A Charleston
B Camden
C Yorktown
D Savannah

29. Under the Articles of Confederation, Congress had no power to
A declare war
B impose taxes
C maintain an army
D regulate foreign trade

So what's the alternative? Luckily, we don't have to reinvent the wheel. Various former tests (most of which have fallen out of use because standardized MC tests are logistically simpler) provide a blueprint pointing towards the type of assessment we should want. Take the Maryland School Performance Assessment Program (MSPAP) which was in place throughout the 1990s. These tended to be weeklong interdisciplinary assessments which drew on a combination of short-answer, extended-response, a little MC and some group work to really provide a more holistic picture. For instance, take the start to this 1997 8th grade science assessment:

You belong to a team of scientists working at the Goddard Space Flight Center in Greenbelt, Maryland. For years, you have been collecting information about our solar system sent to you by a Voyager space probe. Recently, the probe has traveled out of our solar system. Imagine that Voyager is sending back data about an entirely new solar system. Here are some basic facts
about this new solar system:
• There are four planets revolving around a central star.
• For easy reference, the planets have been assigned symbols instead of names (triangle, x, circle, square).
• The planets have nearly circular orbits that do not overlap.

The table below and the diagram on page 30 of your Student Response Book show the information we currently have about the new planets in this new solar system. Study the charts to look for patterns in the data and to decide what additional information you would like to have to describe this solar system and its planets as completely as possible.

[table with data (tilt of axis, surface temp, number of moons, composition) and a diagram of the planets orbits]

Part A: Describe a pattern in this data which is similar to data from our solar system.
Part B: On the lines below, write AT LEAST three questions (more if you can) about additional scientific data on the new solar system and its planets that could be gathered with further study.

2. Working with the members of your group, use the data in Table II and the “Orbit Diagram” to complete the “Orbit Data Logs” on page 33. Draw in the symbols of the planets in their correct positions and order for January, February, March, and April. You may use the “Orbit Data Logs” worksheet to prepare a first draft of these drawings and to make certain that all members of your group agree on the answer. Then copy your group’s answers onto each of your “Orbit Data Logs” on page 33.

Later, questions include asking the students to think about how predictions they made in one part of the exam related to predictions they made earlier and to "Write a description of how your ideas or the ideas of others were influenced by working in a group."

I'm not going to claim that the MSPAP is a perfect exam by any strech of the imagination, but just consider the contrast between it and the SOL! The MSPAP requires active engagement of far more rigorous thinking skills and would be nearly impossible to pass without both scientific knowledge and knowledge of scientific thinking. It can also align nicely to a curriculum which is designed to teach those skills.

The downside of tests like the MSPAP is primarily logistic. They take longer to administer, they cost more, they take longer to score and the scoring is less "objective." Yet if logistics are the controlling factor keeping us from implementing a far superior system which will improve pedagogy and student learning, then logistics are a pill worth swallowing.

There are better tests and worse tests. The overwhelming bulk of state assessments fall into the latter category. What exams like the MSPAP show us, however, is that there is not an innate disconnect between teaching critical thinking and assessment/accountability. It's just a disconnect that currently exists and desparately needs to be addressed.

I'm really starting to wonder about Lori Aritani: In today's Washington Post, she gives us another weak article which is rife with stereotypes about gender roles.

For almost three decades, high school students in Montgomery County have built homes as part of a construction and trades program based at Thomas Edison High School. This year, for the first time, design students have a hand in the process.

Courtney and more than 100 of her classmates will compete to pick colors for the shutters and siding. They will decide whether the family room should be Early American, Country French or even jailbird Martha. And they will recommend a color scheme to make a soak in the second-floor spa tub that much more soothing.

In contrast to students in the carpentry program -- a testosterone-fueled group of karate-kicking teenage boys who have spent the past four months hammering together a 2,252-square-foot house -- most of Courtney's classmates in the interior design program are girls with penchants for fur-trimmed parkas and Juicy Couture bags.

Just stop for a moment and ponder that last paragraph. As my friend put it, "What year is this, 1955?" I'm assuming that, what, the girls rush home to bake dinner before the hard-working boys get back from doing their hard labor?

The worst thing about this article, however, is not even the abhorrent gender stereotypes; it's the missed opportunity. This is a story about what goes on inside classrooms, which is great, but we learn very little about the context. Where's the discussion about the pluses and minuses of vocational ed.; do these projects cause increased achievement in math, for example, or are there other externalities; what exactly is the "construction and trade" program; is this sort of program common?

The only things that you learn from the article are that this program exists, students seem to enjoy it -- oh, and girls don't like getting their hands dirty.