Where the light is

Repeating a joke:

Late one night, a cop came upon a drunk, on his hands and knees, under a streetlight.

"Hey buddy," the cop asked. "What're you doing?"

"I'm looking for my keys," the drunk replied.

"Where did you lose them?" the cop asked.

"Over there," the drunk replied, pointing down a dark alley.

"Then why are you looking over here?" the cop asked.

"Because this is where the light is."

Ron Huberman, the CEO of Chicago Public Schools, has made "performance management" a central tenet of his management strategy. The key metric for measuring performance is standardized test scores. Because that is where the light is.

The impulse behind performance management I think generally is okay -- what one does in the world should be based on what one observes and experiences, and not just what one cooks up based on whim. That is the best part of science. Where this falls down is when one starts confusing the observed quantified data for observed reality.

Steve Talbott, whose work I tremendously appreciate, has written quite a bit about the deep hole that a lot of modern scientific practice has fallen into. (A good collection of his writings are available on the From Mechanism to a Science of Qualities page). In "The Vanishing World Machine", Talbott argues that modern science began when people turned away from labyrinthine abstractions along the lines of "how many angels can dance on the head of a pin" to observing the world around them. But science, especially of late, has made something of a U-turn.

It's true that we no longer ask, 'How many angels can dance on the head of a pin?' but we are certainly entranced with the question, 'How much data can we store on the head of a pin?' And our trance is only deepened when the answer turns out to be: 'a hell of a lot'. What many haven't realized yet is how easily our preoccupation with the invisible constructions on that pinhead blind us to the world we originally set out to perceive and understand in its full material glory.

The world ceases to be the world of dandelions and earthworms and snowflakes and becomes instead a mass of numeric data. Talbott asserts that "our preoccupation with workable mechanisms, far from contradicting our preference for abstract cerebration, is itself a primary symptom of our flight into abstraction and our refusal to see the world." Much of modern science has turned away from the dark alley of the world, back to the sterile streetlight of abstraction.

In modern education, the turning away takes the form of poring over reports of easy-to-process (thanks to computers, etc.) multiple-choice test data. The living child has been replaced with a data shadow. And teachers are pushed into interacting with the shadow, not the child.

Setting aside consideration of the political forces behind the overdrive to privatize education in Chicago, and looking at Huberman's emphasis on performance management from a strictly methodological perspective, I argue that it rests on a broken understanding of science and a disturbed view of the world.

What might be an alternative? I quote from Talbott again, this time from a chapter in his most recent book, Devices of the Soul. Talbott quotes a description of the assessment process at a Camphill community, a network of residential communities for people with special needs.

Every week the staff of a house or entire facility come together to discuss a particular child:

The child's case history is read, and then the teachers, helpers and nurses give their reports and impressions of the child in question. Many symptoms, signs and features are collected until -- usually under the guidance of one of the doctors -- the image of the child arises. His habits, achievements, faults and failures are laid out in such a way that gradually a complete picture of his individuality appears.

In this picture the staff find guidance that enables them to clear a path for the child's continued growth.

Talbott then comments:

The contrast with the mentality behind standardized testing could hardly be greater. Certainly teachers must assess student performance -- and in the most profound and intimate way possible. The problem with standardized testing is that it avoids any such rigorous assessment. It is a hopelessly crude tool, a means of studied ignorance rather than deep understanding. And, as a side effect, it removes all flexibility, the living qualities, from classroom engagement. When you know in advance exactly what knowledge the student-container is supposed to hold, there's not much incentive to attend to the particular gifts and developmental needs, or the consuming interests, of the individual learner. Standardized testing is not student assessment; it is the refusal to assess.

Let me repeat that last sentence: "Standardized testing is not student assessment; it is the refusal to assess."

The keys to making education work aren't under the convenience of multiple choice testing. They are somewhere in the dark alley. Although, when you think about it, it really isn't that dark -- there is no shortage of humane, practical, working strategies to making education work.

jd

I attempt to develop something with Etoys

I just finished a project for a class in multimedia applications. I used Etoys, the Squeak Smalltalk environment, for the development platform. Here is some reflection of my experience...

Etoys works best, like most things, when it is used for what it was designed for. Etoys is something of a desktop lab for exploring concepts -- check out what's on the shelf, see what happens when you mix things, hope not to blow anything up. The underlying pedagogical theme is Seymour Papert's constructionism (a play on, and an extension of, constructivism). Kids can learn by building things and trying them out and figuring out how to make them better.

So I can see using Etoys in scenarios where students are assigned a task, like "create a virtual aquarium containing virtual fish that swim around", "make a car that moves around a track" or even "display the measure of the angles in a triangle using the tools available in Etoys." In the course of constructing the goal, or exploring a concept, all sorts of learning areas are hit, including problem solving, logic, computer science, math, modeling and design.

These are not strictly multimedia applications though, at least as I understand it. To me, an application is a closed experience. The designer has specific learning goals in mind, and creates a series of paths for the user to follow, with different experiences along the way, using multiple media, that will help the user achieve the learning goals. So multimedia applications are limited by the imagination of the designer and the abilities of the software tools in creating the experiences the learner can have via the application. This is like the difference between a webquest and true open-ended inquiry. How heavily programmed is the user experience?

Trying to create a multimedia application using Etoys was ummm a challenge. I wanted to create an environment for the user that anticipated errors and provided enough feedback (which seems like 80% or more of application development). I wanted to guide the user through stages of the activity. I wanted to program the experience. And for this, I found Etoys to be a clumsy environment, dragging tiles around and opening and closing property views and right-clicking to see halos and such.

My experience was challenged by the absence, as far as I can tell, of complete documentation on the different objects available in the Etoys catalog. (I don't know how much time I spent trying to figure out how to calculate the measure of an angle!) There are various help balloons, quick start guides, and there is free documentation on the underlying Squeak Smalltalk environment and language. But the in-between documentation of various properties and what they mean and how to use them is missing (again, as far as I can tell). There is a fairly active user community, spread across forums, IRC and a mailing list. I had trouble joining the forum, the first place I usually go to find out things. I only figured out I could use IRC late in the process, and hadn't really thought of the mailing list as an option, but that might have worked as well. So support is definitely out there, but better documentation would have served me better. I could perhaps have delved deeper into the mysteries of Squeak Smalltalk to really get under the hood, but quickly getting up to speed with a new language, with the peculiarities of new syntax and operators and precedence and all of that was beyond the investment I wanted to make. There was no Goldilocks solution for me.

But again, I was trying to do something with Etoys (create a multimedia application) that I don't think it was designed for (namely, being a desktop lab for constructionist experiences). I was able to accomplish something, but my product really isn't in the spirit of constructionist learning. My foray has some opportunities for student discovery, but not a whole lot.

But... I could not have figured out the proper niche of Etoys in the classroom except by taking the tortuous route through application development using Etoys. Creating my application (zipped, with a video used in the project) turned out to be my own constructionist learning experience. Now I have an idea of where I might want to use Etoys in the classroom. And while I knew that, as far as computer learning goes, I learn best by having a practical problem that drives me to learn the tools at hand, now I have a name for that learning model: constructionism.

jd

Note: Etoys is free and available for download from the Etoys download page. You can find my project in the Etoys Showcase at this url. I think you need to have Etoys installed on your machine to actually view the project though.

The 2009 Bracey Report

I saw a reference to this in an ISTE Special Interest Group on Digital Equity mailing:

The just-published The Bracey Report on the Condition of Public Education, 2009, by Gerald Bracey, looks at what the author considers to be "three of the most important assumptions about how to reform public education", namely:

1. High-quality schools can eliminate the achievement gap between whites and minorities.
2. Mayoral control of public schools is an improvement over the more common elected board governance systems.
3. Higher standards will improve the performance of public schools.

Exactly! Today these three points are usually passed on as if they are facts. Self-evident truths. Obvious. Of course. They are the bedrock of education policy in Chicago Public Schools.

And of course they aren't "facts", they aren't self-evident; and they aren't obvious. As assumptions, they are fundamentally ideological positions and essentially political.

Some points from the report that I would like to highlight:

• Poverty is an objective factor in educational performance. Poverty has biological, social and psychological consequences that negatively affect educational performance. (The report details many different aspects of this.) Schools (and certainly not teachers) alone cannot resolve this problem.
• The idea that mayoral control of education helps education is strictly a political assertion, and is not backed up by a serious review of its results in Chicago and New York (the most visible instances of mayoral control).
• Organizing education around standardized, multiple choice tests and viewing education through the narrow slit of test data is antithetical both to educating human beings, and if you need a market rationale, to developing workers for the workplace of today and tomorrow.

I recommend the report. It has a great bibliography, too.

The Bracey Report comes from the Education and the Public Interest Center at the University of Colorado and the Education Policy Research Unit at Arizona State. Gerald Bracey is also the author of Education Hell: Rhetoric vs. Reality. Bracey passed away on October 20, 2009.

jd

Gradebook rant #1 and #2

#1 - The first quarter ("cycle") in CPS closed last night (Friday, 11/6) at midnight. But then CPS goes and locks the online software teachers need to use, so no more first quarter grades can be entered now.

I don't get this. If the quarter ends on Nov. 6 (Friday), the next one can't start until Nov. 9 (Monday). And report card pickup this year isn't until 11/18 or 19 depending (due I think to the late school year start because of the late Labor Day this year, and Nov. 11 Veteran's Day getting in the way). So why lock Gradebook at the cycle end and prevent teachers getting last minute grades in? Not that I want to spend the weekend getting final quarter grades in, but I'd rather do it clear-headed on a Saturday than late Friday night. Why not lock it Sunday midnight?

There is an implicit understanding that teachers cannot complete their work within the standard 6.25 hour CPS workday (hence closing Gradebook at midnight and not the end of the workday). The Professional Development Day (yesterday, Nov. 6) as a day to get grades in is a joke, because it gets loaded up with, well, PD. So what is CPS thinking? Most of the teachers at my school were assuming they had some time to get the quarter grades in, and only heard Friday (yesterday) morning that the deadline was midnight last night. I'm guessing more than 3/4 of teachers will need to have their Gradebooks unlocked by the school's Gradebook administrator or heaven forbid, someone in the technology support area.

It is another example of trying to force (what looks to me to be) silly and not-thought out dictates and demands on the troops-in-the-trenches (and believe me, it is WWI out here), which ends up making more work for everybody all the way round.

I guess the good news is that the time I was going to spend today fixing up the grades is freed up now.

#2 - I attended the CPS Information Technology Services (ITS) "TechTalk 2009" event last Tuesday at the UIC Forum. TechTalk is an opportunity for the technology apparatus at CPS to talk to the "techcos" (a conflation of "technology coordinator") from the 600 CPS public schools. [Aside -- the techco position is a role, not a position. Typically a teacher or some other staff member is assigned the role of "techco", not hired as such. Techcos form a critical rubber-hits-the-road role in the chain of CPS technology service delivery, but they are stretched between multiple responsibilities. I tire myself thinking about this.] Anyway, at TechTalk 2009, attendees were treated to a talk about Gradebook, which is CPS's branding of a commercial third-party product called Gradespeed.

The presenter gave one of those classic tech talks where the end users were morons who did not read instructions. In fact users are generally smart people confronted with a confusing interface that could easily be fixed up. It is a poor technologist who blames the user for design shortcomings. Case in point: teachers can enter a numeric code to represent comments like "Is too easily distracted" (but only one for elementary schools we learned, even though there are five boxes!) . If the code is "028", you need to enter the leading zero. Why? Lazy programming. The person I was sitting next to and I agreed that this could be fixed in a couple of lines of code and eliminate untold headaches on the part of teachers who enter "28" instead of "028". Thinking about it, it isn't a couple lines of code -- it is less, just a matter of wrapping the user input in a couple of functions that check the input and clean it up for the user -- maybe 20 extra characters of code.


Example #2: The user has a row of links at the top of the screen to control options:

This is an online application, so you might think that you can click on the image, that it is a button? Nope. Only the text is a link. I have worked with computers for over 25 years and this one tripped me up. One gets in the habit of thinking how things should work, primarily a habit of life experience. Software designers design an experience, and the best ones anticipate user expectations. The user should be at the center of the design, not the designer. That's what "participatory design" and incorporating real live users in the design process is all about. Put them in a room and see what they try to do before releasing the product.

Another example is that when entering assignments, you must click an "Add" button to add a new assignment. However, after clicking "Add", you see a blank form again. No program feedback is provided to indicate that the assignment was saved. Was it saved? How would you know? Only by exiting the "Add" function by clicking "Finish" (what are you finishing?), and seeing if it now appears in the assignment list. Again, some simple code to acknowledge that the "Add" operation was successful would be a trivial enhancement.

The computer experience is often akin to stumbling around with a bucket on your head -- you have precious little feedback on where you have been or where you are or what lies ahead. The best software designers provide lots of feedback: cookie trails with trackback links, alerts, the use of color in logical places, and so on. Not so with Gradespeed / Gradebook. Instead the poor user who only wants to accomplish a task and not "work a computer" is blamed for not understanding the software.

Too much ranting? It's a beautiful day today. And now no grades to enter. Yay I think.

jd

Charter school limit raised

I was reading through the Chicago Teachers Union newspaper (why do they print a four-color paper on clay-coated stock, especially given the financial problems the union is having?), and saw an article on Illinois Senate Bill 612, which was signed into law last July. Some highlights of the bill according to the article ("Charter school law includes wins and losses", p. 6):

• Doubles the cap on charter schools in the Chicago area from 30 to 60 single-campus schools, plus an allowance for up to five multi-campus charters targeting drop-outs.
• 75% of charter school teachers across the state will need to be certified, up from 50% in Chicago before.
• Charters need to disaggregate multi-campus data (single charters running multiple campuses was
  
• There can be 30 contract schools (bound by the Illinois School Code, but the CPS Board contracts out management of a school -- click here for more on the types of schools under Renaissance 2010), plus five additional contract "turnaround schools"; but the contracts go through a new authorization process.
• A parallel bill (SB 1984) says that charters fall under the Illinois Education Labor Relations Act, clarifying that teachers can organize at charters and contract schools.

The thing that sticks out the most to me is the doubling of the charter school number. In practice this will mean the closing of 30 more public schools because after students fill up the charters, the remaining public school enrollment will not justify keeping the school that lost students open.

In other CTU news, Jay Rehak and Lois Ashford won trustee seats on the Chicago Teachers Pension Fund in Friday's (10/30) election, defeating the incumbents supported by the CTU Executive Board. Rehak and Ashford are members of the Caucus of Rank and Educators (CORE). Per their web site, "We hope to democratize the Chicago Teacher’s Union and turn it into an organization that fights on behalf of its members and the students we teach."

This sounds like a rebuke of the current union leadership. What is the CTU doing anyway? And why does the CTU president draw two six-figure salaries? How can she do both jobs effectively? Which means she is doing neither one effectively. The new charter school bill, the school closings, Huberman's rampage, etc. etc. as cases in point...

jd

Creativity and education

What is education for? This is a the question explored in Daniel Wolff's new book, How Lincoln Learned to Read. I confess I haven't read the book yet, but I did hear him speak at a book signing last week, and it is in the queue. This question -- what do we want out of education, for ourselves, for our children, for our community ? -- seems to fall into two broad camps. One camp (and where I hang out, or try to) sees education as a self-maximization process, the be-all-you-can-be approach. Economics is secondary, and will follow in one way of another -- more of a statement of faith I suppose, but...

You can identify people from the other camp because the words "economic competitiveness" will likely pop up early on. Education is an economic function, in Marxist terms, the social reproduction of labor power. To be economically competitive in the global market, you need to know (fill in the blank). Pablum about the nature of work today, what employers look for or need to be competitive is trotted out, and then blueprints for "education reform" soon follow. Case in point: Thomas Friedman's October 20 column in the New York Times titled The New Untouchables. The "new untouchables" are the workers who bring something special (dare I say, something human?) to the workplace, like creativity or interpersonal skills, and thus are untouchable (i.e. difficult to replace) in the workplace. This has been a common theme over the past 25 years at least -- jobs that can be done by a machine or by a cheaper counterpart elsewhere will eventually be done by a machine or moved elsewhere. Within the dismal terms of the global economy, Capital will seek out the greatest return with the lowest cost, and Labor will always be an important front in the war to maximize profit. What is difficult to replicate in technology are attributes like (per Friedman) entrepreneurship, innovation and creativity.

The two camps do overlap in some areas. Friedman holds that creativity is a key feature of the successful modern worker; a humanist educator would agree that creativity is a very special human potential and should be nurtured. "Entrepreneurship" is a narrow economic term and situates education in terms of the marketplace; a broader conception is to foster initiative, forward thinking, exploration, fascination and a willingness to try things out. Nevertheless, there are common elements in both camps.

The sad irony in this is that both camps are calling for a radically different kind of education than that being forced down the throats of teachers in Chicago and elsewhere. Data-driven instruction may sound scientific and efficient, but at the heart of it, it is antithetical to the kinds of skills Friedman is writing about that today's economy needs. (And Friedman has narrowly conceived of the human being as only a worker -- it doesn't begin to touch on the full range of human possibility.)

Simply put, multiple choice is antithetical to creative thinking. In terms of either globalization (Friedman) or humanism, the approach is wrong and destructive. Our society and planet face big big problems, and time is to critical to be wasted screwing around with wrong solutions.

jd

Huberman's vision for education in Chicago

I heard the Chicago Public Schools Chief Executive Office Ron Huberman speak on September 29 to a group of principals and teachers where he once again laid out his agenda for Chicago public schools.

Huberman's plan is organized around themes of "performance management" and "data driven instruction." Like other Mayor Daley appointees, Huberman (who, for the sake of possible non-Chicago readers, came to the school system from the Chicago Transit Authority) in theory brings private-sector style managerial expertise to the sprawling bureaucracy of CPS. George Schmidt, in his front page story ("Data Driven Drivel", sprawling in its own way) in the latest issue of Substance, uses the term "narrative" to describe Huberman's pitch. The term is well chosen I think. "Narrative" captures the essential point that Huberman is telling a story about what is happening. In this case he is framing it within the too familiar story of capitalism and the market, with all of the narrow and limiting assumptions and possibilities that that dismal story allows: In his narrative-story-vision, schools are education finishing plants that add education "value" to children. Value-added is measured by the change in standardized test scores over time. Individual schools are the education providers (not the school system, which only provides infrastructure), and schools, in the Obama-Duncan-Huberman world of "choice" (another marketplace term), compete with each other for raw material to finish. In this story, parents are free to take their raw material to whichever finishing plant is going to add the most value.

Hubeman is very explicit about this. In his September talk, he described each child as having a "backpack" of money, the money collected from taxpayers to educate children. CPS collects about $10,000 per child per year, or $14,000 for Title I schools (schools with large numbers of students from poor families that receive additional Federal money). In Huberman's vision, parents should be able to take that backpack of cash to the school of their choice. Schools will have consumer report cards showing their "value-added" scores to help parents in choosing the schools most successful at raising test scores. Teachers at each school will compete with the teachers at other schools in the city, both public and charter, and through that marketplace competition, their schools, at least in theory, will become better finishing plants.

So what is wrong with Huberman's story? I think that in practice, schools doing well now will catch the best students, and struggling schools will fall further behind. It is the same process of polarization that plays out in the economy at large -- the rich get richer, etc. Likewise, charters skim off the best students, or the students with the most parent involvement in their lives (and there is some correlation there). Public neighborhood schools become the schools of last resort -- they cannot refuse to take a student in their attendance boundaries. The students with the most needs will pool up in the public schools. Their test scores will sink the neighborhood schools, feeding a vicious cycle of increasing privatization. More charters will spring up to snatch the backpack of cash; the teacher's union will be effectively broken; and the concept of public education as a great project of Democracy will die.

Management by FUD -- Fear, Uncertainty and Doubt -- is the order of the day in many CPS schools now as the pace of this process accelerates. It is questionable that that strategy works in private companies; in a school it is disastrous. Teachers feel like they are under siege. They have a difficult-enough time as it is, coping with classes with too many students, with too few resources, no recess, and a steady stream of new mandates from on high. The axe of school closing and unemployment, based on one testing event a year, now swings over their heads as well.

The marketplace model has other problems too. Certainly accountability is important, but it needs to recognize that schools are not separate from social and historic forces playing out outside of the school building. But reducing accountability to test scores is fundamentally flawed. Test scores are a convenient metric, and fit in with the overall quantification of everything that is part and parcel of the marketplace metaphysic. More on that in my next post.

jd