by Alain Jehlen in Schools and Youth
Posted on November 8, 2010 at 11:21 pm
Last Modified on November 13, 2010 at 12:02 pm
loading…
Here’s a radical idea for the Somerville schools: Stop trying to boost MCAS scores. Except in high school.
I don’t mean stop teaching, but let’s turn off the laser focus on raising scores.
An example of the difference:
Somerville teachers are being encouraged to put up “data walls” in their classrooms with each child’s test score, or score growth, or some sort of class ranking of scores, represented on a chart. The children are not identified by their real names, but they all know where they are on the wall.
The low kid in the class sees him or herself at the bottom every day and the top kid sees him or herself at the top. That’s bad for both of them. And it’s teaching them nothing except the false lesson that nothing matters in life but your ability to tackle a standardized test.
The intention is to motivate students to score higher so our schools will make “Adequate Yearly Progress” under No Child Left Behind, but:
1: That’s a hopeless quest.
2: It’s not helping our kids develop their academic skills or anything else that we want from an education.
Let’s start with 1: It’s hopeless.
No school in the city is making the No Child Left Behind’s “Adequate Yearly Progress” except the Brown, which has by far the fewest low-income students. (The charter school didn’t make it.) See for yourself on the Department of Education’s web site.
Somerville schools can never catch up because NCLB’s standards keep rising. If a school somehow did manage to make the standard one year, it would soon flunk again. Even the Brown is doomed.
We’re no different from the rest of the state and the country. Everywhere, more and more schools are failing this impossible test, and low-income schools are the first to fail. In Massachusetts, 57 percent of all schools failed this year.
By 2014, NCLB says every American student must be “proficient” in language arts and math. That year, nearly every school in America will fail because no nation has ever come close to that standard, unless we set a very low level for what we mean by “proficient.” Some states have tried to help their schools pass by dumbing down their definitions of “proficient,” but who are they fooling?
Congress will probably change the law before everybody flunks. Let’s hope the new version bears some relation to reality.
Meanwhile, we have teachers who want to teach and kids who want to, and need to, learn. Why waste their time and turn them off to school with test prep and score competition?
Which brings us to 2: It’s not helping our kids.
If test prep helped kids learn valuable skills, it might be worth the pain. But we have scientific proof that it doesn’t. Prepping for a standardized test—reviewing sample questions, analyzing the kinds of questions likely to appear, staging competitions and pep rallies—these strategies are being carried out all over the America, especially in schools with large numbers of low-income kids. This experiment has been going on since NCLB passed in 2002.
And the results are in!
There’s a national standardized test that’s given to samples of students across the country called the National Assessment of Educational Progress (NAEP). Nobody preps for this test because there are no consequences for low scores. Looking at NAEP scores over the past decade, you would never know that anything was changing in our schools. Here’s an article that shows the NAEP trends and goes into more depth.
Even before NCLB, national studies showed high-stakes testing has little or no effect on student achievement. Those studies were ignored by Congress in their desire to “do something” about students who don’t learn enough in school—preferably something cheap.
But test scores on many state tests, including MCAS, have gone up. Doesn’t that mean kids are learning more? No, it doesn’t.
Scores on the state tests are rising because those scores determine success or failure under NCLB. Schools are putting massive efforts into boosting scores on those tests. The NAEP results show that the improvement doesn’t even carry over to a different standardized test, never mind to real life situations. The emperor has no clothes.
(Massachusetts has climbed to the top on the NAEP, but probably not because of MCAS. The rise in Massachusetts is small—we were always near the summit. And other states don’t show even that modest improvement on the NAEP, although their state test scores have gone up, too.)
It turns out that when you attach major consequences to scores on a particular test, that test stops being a valid measure of achievement. This phenomenon is well accepted among testing experts and even has its own name: Campbell’s Law. All across the country, schools are ramping up test prep. Scores go up, but real achievement doesn’t.
That may protect the grown-ups, but it doesn’t help the kids.
The situation in our high school is different. Until the state policy changes, students will have to pass the high school MCAS to get a diploma, and they do need that piece of paper, so there’s a reason to help them pass MCAS even if test prep doesn’t improve their skills.
In elementary school, let’s do everything we can to help kids learn and love learning. But when it comes to test prep, let’s just say, “No!”
loading…
Doesn’t failure to teach to the test, and the resulting drop in scores, mean the teacher gets bumped down the tenure list or the school loses funding?
This doesn’t mean I have a good alternative idea that any but the very well-to-do can afford, but the fear of getting fired and having even MORE kids per class (and fewer books, computers, musical instruments, sports coaches, etc.) could be significant stimuli.
loading...
First, at least in the current contract, there is no connection between teacher tenure, promotion, or income and test scores. That is a scheme being promoted by Secretary Duncan, with some kind of fuzzy and insubstantial support by the Massachusetts Department of Education and the Governor. It is being aggressively challenged – by both research and by the national and state teachers’ unions. While the ultimate decision is still unclear, it clearly does not have impact on any existing teacher or teacher contract in this state.
Second, I actually disagree with Alain, but not completely. While tests most surely do not indicate a high probability of future success – neither in school nor in a job – for most fields and at most levels, they do indicate relative skills at test taking. That itself has some, modest but real value. We might question why the state spends $50,000,000 a year on a test that takes months to give results (and, for particularly the lower grades, a large portion of kids’ life to give feedback!), but the idea of a multiple choice test as one of several measures of student skills is still quite valid. And it’s also the law (a handy coincidence, at least).
What is missing are the rest of those measures. And they do exist. They include homework, attendance, grades, and other tests; projects, participation, and creative contributions; engagement, maturity, and other achievements both visible and attestable (in every way from a paper to a dance, from a teacher or employer recommendation to a feat on the football field).
Also missing are the metrics – the standards – for such achievements, although they are well known and tested and documented. Most of those measures were established in the 1980′s and 1990′s, and summarized in what was called the SCANS Report (Secretary’s Commission on Necessary Skills, from the Department of Labor in 1992). The author of that report, and national leader in “soft skills” evaluation and analysis, is Dr. Arnold Packer, who is consulting on a pilot project already under way at Somerville High School.
Those skills include “responsibility, teamwork, listening, acquiring & evaluating information, interpreting information, creativity, negotiation, and working with cultural diversity.” Some systems add a few terms, and some – like the one at Tufts, for example – consolidate a few. The Tufts “soft skills” assessment of applicants targets “wisdom, intelligence, creativity, and the synergy to coordinate these skills.” The point is that these terms are deliberately subjective, and have traditionally been dismissed. Yet they can be demonstrated in portfolios. And those portfolios ARE mandated by law and kept in and by the school system. So the real challenge is to teach students to evaluate themselves and others, and to evaluate those portfolios against reasonable standards.
For decades we have known the skills, but ignored the problem of their subjectivity. Yet, if we actively pursue those skills, they are less subjective than deciding which multiple choice question is more important or how to score an essay exam!
For example, if we want to “score” responsibility we look at on-time performance, at perseverance and attention to self and others, and at concentration even when doing an unpleasant task. It’s not hard to see such factors, nor to score them; nor – and this is the critical issue – to reflect on those scores with teachers, parents, and the kids themselves. They may be subjective, but that subjectivity really describes a net of those reflections. Build the net, don’t ignore the skills!
And that is precisely what we are piloting at Somerville high, with, this semester, six teachers and twelve kids, from 9th through 12th grades. Ideally, we’ll continue that pilot with more next semester and more and more in future cycles. It doesn’t have to be “required,” since it is both fun and productive in itself. Kids like it, as do teachers. Ideally employers will agree with those scores, and the portfolios that result will improve kid chances for college, for jobs, and for better careers.
That, after all, is the point of it all.
loading...
Joe – thanks for the highly informative post! I certainly didn’t know there even existed validated instruments for assessing “soft” skills, nor that the punish-the-teacher approach hadn’t breached the walls of Somerville yet – both excellent to hear.
Can I ask you another question? Do you think that high-stakes test prep, or any other factor or combination of factors, crowd out enough of the actual learning that you describe to make it worth lengthening the school year, or doing away with summer vacation entirely? I realize that it would take a lot of economic stress who can’t work when their kids aren’t in school. But from a purely educational standpoint, do you think it would be a good idea? Or does summer break serve an educational end as well?
loading...
(take a lot of economic stress OFF OF PARENTS who can’t work when their kids are in school.)
loading...
I really think – and this isn’t a popular thought – that most curriculum is pretty irrelevant: that test prep is a useful skill since so kids face so many tests in school, college, and grad school, but fairly useless when it comes to solving real problems in real world situations. (The Association for Public Policy Analysis and Management’s conference last week at Boston’s Hyatt had at least four forums on precisely this research, all of which concluded that testing, formal prep and exclusionary screenings have very low impact on real knowledge, and that most – even technical – careers already use just-in-time information better and more effectively than drill-and-practice could ever produce.)
Regarding the school year, I’m ambivalent. I really think that most of our Somerville schools are hotter than hell in the summer, and that air conditioning would be a major public expense, particularly if it were for many students at many sites. I also think we need many more alternative activities for kids – since the calendar was really derived from 19th century work schedules on farms and factories no longer relevant. But I don’t think most of those alternatives are best exploited in schools. The school “factory model” has real, but limited, utility. Instead there should be projects, ways to improve the city, ways to help elders or younger kids, ways to engage in pre-career, pre-work exploration. Some of these ways might be through churches and community groups; some may be through parent groups and teacher and school extensions. Yet the real shame in Somerville education should be charged to Tufts and Lesley, who use our capacities and should be far more active in helping their own communities. As the largest employers with real skills and knowledge related to these needs, their absence is a profound and enduring proof of their pseudo-elite status, and they should be paying – big time – as long as they keep their white gloves off our real problems.
loading...
Re: Tufts and Lesley – does the School Board try to form better connections, or is there a designated liaison? While their respective administrations might consider townies a nuisance, it might be possible to identify individual faculty members who could see win-win possibilities in dirtying their and their students’ hands… (OK, I can be a bit of a Pollyanna, but better to voice it than not!)
Re: the factory model, may have already seen it, but this short video is worth watching and sharing: http://www.youtube.com/watch?v=zDZFcDGpL4U
loading...
Ah, the age old question of how much should the institutions of higher learning “pay” for being part of the neighborhood?
Barbara Rubel, community relations director from Tufts, has an interesting take on how much Tufts currently does here, http://www.greatersomerville.wordpress.com(in the archives). Lesley is a fairly recent addition to the Somerville community and is harder to measure. I’ve suggested in the past that the city have a real “town/gown” entity whose main purpose is to engage, implement and measure the amount of effective “contributions” from these institutions.(monetary and non-monetary)
The past and curent method of leaving these initiatives and measures in the hands of the Mayor, school committee and certain ward alderman has not, in my opinion, been very effective at all.
Especially when it comes to their engagement and contributions to our education system.
loading...
Excellent, thought-provoking article. I was fascinated to read about Campbell’s Law in your article. (Reminds me of Heisenberg’s uncertainty principle!)
If there’s a strong correlation between the socio-economic profile of the student body and MCAS scores, perhaps we could just skip the test and come up with an index based on the number of kids at a school who are receiving free/reduced lunches. A high “score” on the free lunch indicator would richly reward the school with all kinds of wonderful educational resources.
Imagine if we had all the money we’ve spent on military initiatives in the past ten years to use on effective educational resources for our children.
loading...
My understanding is that the new report cards are going to give feedback on student progress in the “soft” areas Joe B. mentions. I think that’s a great goal, and I’m very curious to see the new reports when they come out later this month.
As for “teaching to the test,” I honestly haven’t seen much of that at all – a little bit in the spring, which if anything, seems designed to calm the kids and give them confidence to face what some of them see as a huge ordeal. Then again, my child is still young and has only been through one MCAS cycle so far – maybe the test prep ramps up a lot in the higher grades. If so, I would worry about what is lost when so much time is devoted to MCAS prep. What I’m seeing now is that teachers are working really hard and coming up with projects that engage kids all year long and that follow the Mass Curriuclum Frameworks. I’m seen real improvement in my child’s reading and comprehension skills. I would really hate for the productive, valuable work of teachers and students to take a back seat to achieving higher test scores.
loading...
In Japan, children go to “cram school” from an early age. Getting into a good high school requires passing a standardized test. So does getting into a good college. The college entry test requires history, literature, math, science, and English. However, the teaching methods are mostly rote memorization (as one would expect when teaching to a test), with the result, for instance, that students may pass the English exam without being able to actually speak English.
Japan tried backing away from its emphasis on standardized tests about 10 years ago, and discovered that its test scores dropped. There’s now talk about reemphasizing them.
I have mixed feelings about this issue. I was horrified a number of years ago when, as a teaching assistant in a U. Mass. programming for poets course, I heard the students complaining about having to do grammar-school arithmetic during a quiz. And that, although the students talked a lot about the “class average”, they actually didn’t know how to calculate one.
This is the 21st century, and we live in a fast-moving technological society. Maybe standardized tests aren’t the answer, but we as a society are not going to survive if we don’t get educated and stay educated. Societies that value learning will eat our lunch.
loading...
And Japan’s recession, now into its second decade, suggests that that lunch won’t be sushi.
loading...
One way that Lesley gives back is to give an $8,000 per year (total $32,000) scholarship to any graduate of Somerville High School.
loading...
Alain, I agree with your arguments against pushing students to score highly on the MCAS, to the detriment of other educational pursuits.
I wonder if alternative measures, such as the rates of college enrollment, college scholarships, and job placement following graduation would be better than looking at test scores? At least they seem more practical, and less prone to contrivances such as test prep.
fyi… State Rep. Carl Sciortino has already taken legislative steps to reduce the reliance on the MCAS… and Gov. Patrick’s administration endorsed national standards that could lead to the replacement of the MCAS.
loading...
Barry’s suggestion of other “outcome measures” is both timely and locally important. Last year Superintendent Pirantozzi sponsored a pool for who could guess most accurately how much financial aid the Class of 2010 actually earned. Ironically, as college prices increase to $200,000 for a private four year degree, the 350 Somerville graduates represent an “opportunity cost” to measure a “return on investment” a lot more concrete than some MCAS average. While all won’t go four years, nor all in private colleges, many, many more would, should, and might go more, further, and with better funding if the system – and the larger community – invests the time and money to make it happen. And that is both a short term bonus and a huge long term gain for Somerville families since a BA is worth well over $300,000 more in lifetime earnings than a diploma.
We might actually find the real total of those scholarship and loans if we hold a contest to let kids share their “notice of grant award” letters privately, and give prizes for 2 year, 4 year, public, private, and trade schools….
Don’t count on the Patrick administration to reduce their dependence on tests, incidentally, since those national standards would replace MCAS with an even more expensive and extended test cycle. The state’s recent leadership of a Race to the Top Assessment consortium may have some effect, but I wouldn’t count on it being positive – since both Patrick and Obama seem wedded to some pretty regressive testing patterns.
Part of the problem of tests is that they are typically framed as “yes or no,” whereas tests are fine as long as that’s not all there is. And that’s what has provoked most of my loquacity regarding Alain’s initial post. Don’t argue pro or con MCAS. Instead, formalize some other measures and give that MCAS data some perspective. If there were a little more to go on, we might not worry so much about what little there is in the first place.
loading...
What a fabulous post! I hope some Villen teachers will weigh in here- anonymously, if necessary.
Does Teacher development fit in this discussion?
loading...
I also posted this to the Somerville-4-Schools yahoo group and several people there have commented. Here’s some of that exchange:
Nov. 9
…I can see where data walls and test prep can be counter-productive if taken to their logical extreme.
What I didn’t hear in your posting (and article in NEA journal) is what you suggest as an alternative?
-Yes, NCLB has some problems but I don’t think that ‘not testing’ is a better alternative.
-If you do test, it also seems logical to take some action based on the test results and,
-it also seems reasonable to have Yearly Progress as an objective, especially if our system currently isn’t producing the desired results.
I think that the problem is that NCLB sets up an explicit competition for scarce resources, thus creating the incentive for ‘test prep’. I’m sure that we will agree that more can and should be spent on education, but resources will always be scarce. The operative question isn’t whether NCLB has specific problems, but rather whether NCLB is better than the old method and/or any other proposed alternatives. I don’t know the answer to this, but the few models/anecdotes that I’ve seen for alternate resource allocation don’t strike me as being any better; NH using only local property taxes to fund schools and Newton exploiting another bad allocation model to build a $200M High School using State funds. At least with NCLB the funding decisions seem to be based on educational outcome instead of affluence or political power.
The NAEP data cited in your article does give some pause to the NCLB approach, but I don’t yet see this as definitive as there are many other socio-economic trends that transpired during the two periods studied. Could it be that this difference in educational outcome can be explained by the growth in wealth inequity that occurred during the reign of NCLB and, were it not in place, the educational gap would have broadened? Correlation does not prove causation.
It seems that another way to look at this is, prior to testing, the general assumption was that the Western suburbs had great schools and Somerville did not. I believe that MCAS test data has mostly debunked this myth (supported by 2002 S&P/School Matters report) This is even more compelling when you consider that Somerville likely spends much less time/money on test prep relative to these other schools. I believe that, at least for Somerville, this data-centric approach is creating a virtuous cycle that help convince middle-class families to send their kids to our public schools rather than flee to the suburbs where the perceived higher performance is really just a proxy for higher income.
I agree that test prep, data walls and test rallies probably aren’t the best use of time and resources, but I also disagree with the use of the words ‘hopless’ and ‘doomed’ to describe, in any way, the Somerville school system as I have experienced it.
–Michael Chiu
Many points to think about. Let me start with Michael’s last comment about using “hopeless” and “doomed” to describe Somerville schools, because I don’t want to leave the wrong impression.
The Brown, from all I hear, is a fine school and will continue to be a fine school even if it is eventually labeled “failing” by NCLB. My granddaughters are getting an excellent education at the Healey. The older one, in third grade, is constantly writing long stories, in school and for homework. Her teacher keeps parents fully informed of what’s happening in their children’s school lives with personal notes that she must be staying up late to write. There are lots of great things happening at many Somerville schools. But they still can’t make “Adequate Yearly Progress.”
Almost every school in the country is doomed to “fail” by 2014 under NCLB. Trying to avoid that kind of “failure” is hopeless—unless the standard is changed.
Next, Michael’s point that the lack of improvement in NAEP scores doesn’t prove NCLB isn’t working—maybe NAEP scores would have fallen without it. That’s possible, but it would be a strange coincidence if whatever was causing scores to fall so perfectly offset the benefits of NCLB that the result was essentially zero.
But there’s other evidence. Before there was NCLB, some states—like Massachusetts—adopted high-stakes graduation tests. Others did not. Several groups of researchers compared what happened in those states during the years before and after the adoption of graduation tests. Some reported small benefits from high-stakes tests. Some saw no benefit or found negative effects. The differences could have been due to the different approaches they used or the different sets of grades and subjects they examined.
The important thing is, nobody found big effects.
So NCLB simply confirmed on a much larger scale (and with much more damage to our education system) what was already known: high-stakes testing just doesn’t work.
So what’s the alternative? A very good and fair question. For another night.
Nov. 10
I want to at least take a stab at responding to Michael’s comment: If not NCLB, what’s the alternative?
I think he means, how else can America improve academic achievement among groups that aren’t doing well in school today, namely low-income kids, especially minority, and also kids with disabilities. One of the few good things about NCLB is that it requires separate reporting of scores for those groups so they’re not hidden.
I don’t have the answer, but here are some ideas:
First:
Schools can’t solve deep social problems. The high-scoring countries of Europe deal with the effects of poverty by making sure they don’t have so many poor people. They have universal healthcare, more generous unemployment insurance, extra help for families, free pre-school, and many other programs, and the result is that their teachers have fewer kids from desperately poor families. Of course, that costs money.
Second:
Free, high-quality pre-school has been proven by large-scale experiments to make a big difference in the lives of low-income kids. It more than pays for itself. But it has to be high-quality, which means staffed by well-prepared, skilled teachers. Again, many high-scoring countries have that.
Third:
High-scoring countries put a lot more money into helping teachers improve their practice, especially in the first few years, with extensive mentoring, but also later on. Their teachers spend much more of their time preparing lessons and working with other teachers to evaluate and plan their work.
We bring up the rear.
The Organization for Economic Cooperation and Development, made up of the world’s richer countries, puts out an annual report called Education At a Glance. One interesting table shows the number of student contact hours for teachers. American teachers have the most—by far.
In elementary school, American teachers average 1097 contact hours.
The OECD average is 786.
Top-scoring Finland: 677.
The differences are even more extreme in high school.
[http://www.oecd.org/document/52/0,3343,en_2649_39263238_45897844_1_1_1_1,00.html It’s indicator D4]
(Note: This isn’t how many hours kids spend in school. It’s how many hours the average teacher spends with a class. Kids have more than one teacher.)
No Child Left Behind’s approach to schools is like the defunct Soviet Union’s approach to the economy: You don’t trust the locals to make good decisions, so you set numerical targets for everything important.
But the economy is way too complicated for central planning to work, and so is the education of human beings. Nobody really thinks the purpose of school is to produce test scores.
To improve the quality of education, there’s no alternative but to trust the people doing the work. We have to get the best possible people, help them prepare well, improve the conditions they work in, and give them the flexibility they need to do their jobs.
Other ideas?
Alain
loading...
Linda Conte posted this comment to the Somerville-4-Schools yahoo group:
…my own observation [is] that Somerville elementary teachers are rushing to retirement and dropping to teach 2nd or 1st grades in order to escape MCAS with bitter feelings toward the system that has strangled their creativity and taken the fun out of teaching.
MCAS has done damage to our system. Teachers feel controlled by managers who have far less classroom experience than they, who come up with things like MCAS and more tedious data gathering, and “teachable moments” are subverted every day by these outside demands and requirements.
No one thinks either side is all wrong, but certainly teachers feel that no one really asked THEM what could be done better. How can you help your kid learn better? Ask a professional. Ask a teacher. They miss the projects they could delve into without the demands of block scheduling and teaching for MCAS success.
My own kids are strong in math and less so in language arts. I can’t say how all the MCAS focus has effected this outcome, but this English major is a bit grumpy about it.
I’m not against testing that helps teachers know what they’ve managed to get across and helps kids know what they could work harder at, but I AM against tying punitive measures to adverse results that discourage teachers and kids and terrorize those that aren’t fired or dropped out.
Now I’ll stop before I have anything else to drop a preposition for. (Damn.)
-Linda
loading...
There are some very specific pedagogical (!) reasons to discount MCAS which haven’t yet been discussed here. Assessment is for feedback; assessment of a learner is to feed that information back in a timely fashion to improve learning. MCAS takes three to six months to get back to the learners. No wonder people hate it. If you’re 10 years old and they tell you what you “need to know” six months later, you’ve already spent 5% of your life between the test and the grade. That is horrific. It is neither logical nor even practical.
Gaps between assessment and instruction destroy the dialog between students and students, teachers, and parents. And, ironically, the format of the test is designed for rapid response. A multiple choice test is only justified by how quickly it can be scored. There is no serious intellectual foundation to a process of choosing which answer is “right,” particularly on questions – which are becoming more common – where judgment is required. It may be fine to use a multiple choice instrument when answers are a number derived from a linear formula (2 + 2) but it merely compounds subjective judgment when the answers are better or worse, closer or less distant from “truth.” Such questions are often very appropriate, but in the course of instruction rather than evaluation, since you want the opportunity to discuss the question itself as a way to reveal how to derive any of several right answers. When there is a six month delay, the test kills discussion. And, ironically, those are the questions that have the most instructional value.
I spend a lot of time in the schools, particularly in Somerville. I am always impressed at how flexible teachers and students can be; and equally impressed that, when challenged, anyone can slide into a defensive and negative posture. Schools are always at that edge: kids can flip from compliance to defiance; teachers from colleagues to bullies. Sometimes they can flip right back, and use the challenge to grow; but sometimes those postures become habits, habits become standards, and kids’ curiosity dies while teachers’ vitality evaporates. Tests and judgments with insufficient information put rules and curiosity into opposing camps and imply that right answers are the only answers, which is usually a very, very serious mistake.
Finally, these particular tests are immensely expensive for schools and school systems that are seriously under-funded. This is less serious in places like Somerville, where our cost per student is both reasonable and regularly examined. It is a disaster in places like Chelsea and Lawrence, where a substantial portion of the entire year’s budget and activities go into a test. And those are places is where Alain’s international formulae are most oppressive, where the difference between Chelsea and Finland is most extreme.
Somerville is remarkably safe from the worst of MCAS impacts. While all but one of our schools is “low performing,” in fact that means we rarely meet an ever rising standard based on past performance. We have a very, very low rate of MCAS failure and a very responsive network of safety net activities, to intervene as quickly as possible when we get feedback to use that feedback as wisely as possible. We can use the “low performing” label to capture federal and state support to make some dramatic leaps all the more dramatic. But others are neither as wise nor as lucky nor as … progressive. And the dropout rate at their schools is over 50%. Kids vote with their feet.
These extremely negative impacts are not “because of MCAS,” but, rather, because of how some schools use MCAS. Again, those impacts are rare in Somerville, which has a sharply declining rate of failure largely because we have a sharply increasing rate of early intervention to PREVENT failure. We are, in fact and in theory, using tests as wisely as possible. But, again, that is not the case statewide. And is it neither easy nor common.
Meanwhile, it is critical that we create other means of documenting what and how and where students learn, to give context to a very over-simplified test score. Without that context, the tests become the only indicator, and recent studies like NAEP (http://nces.ed.gov/nationsreportcard/) have gross generalizations about skills and knowledge based on vastly over-simplified measures like a single numeric score. Such measures ignore the real vitality of American education, and the massive achievements of good teachers and excellent schools. But that’s another problem. Too bad so many researchers generalize from such poor data. But that’s not really our problem, and MCAS is mostly an inconvenience, not the great dictator of our local curriculum.
loading...
Darn it! Just lost my first response. Take two!
I haven’t read this thread closely, yet.
And I’m aware there is some misinformation in the original post.
The data displays in Somerville schools don’t show how well students have done on MCAS.
The data displays show how much progress students have made on a local, adaptive test administered three times a year. This test, calls MAPS (measuring academic progress) is used to help teachers understand how much progress students are making in individual topics over the course of the year.
Teachers use students’ individual test results to adapt their teaching – both to the whole class and to individual students.
The test is adaptive, which means it shifts based on the individual test taker’s areas of strengths and weaknesses. (No point in continually asking a student questions about algebraic thinking if s/he has already aced the first three questions measuring this.)
The data displays use icons – for example a tableau of birds migrating south where each bird represents a student in the class. The “highest” or “farthest” bird in the tableau is the bird that has made the most progress since the last test period – NOT the bird that has the “highest score.” (Because the test is adaptive, or individualized there is no highest score!!)
I’ll read this more carefully tomorrow, and will share it with the senior management team.
Thanks for the interesting conversation.
loading...
Gretchen,
Thanks for your response!
Yes, the data walls are about MAP (Measures of Academic Progress) scores, not MCAS. MCAS is only once a year. I didn’t say they were MCAS scores but I can see that someone would draw that conclusion.
I’m not sure it makes much difference, though. The data maps are part of the school administration’s effort to get schools to make “adequate yearly progress” under No Child Left Behind.
I looked at several data walls today and all were representations of scores, not score growth, although I understand there are also data walls that show growth instead. I don’t know which is more common.
But even if they were all growth scores, the message to kids is still that the point of school is to do better on a standardized test. It isn’t. And I don’t really think anybody in the administration thinks it is.
I’m all for using MAP if teachers find that it helps them diagnose learning problems. But the data walls aren’t to help teachers diagnose problems. Their purpose is to motivate kids to work harder so they’ll score higher on the next test.
If focusing on test scores actually improved student skills, we could have a debate as to whether the improvement was worth it. But I want to repeat the point I made earlier. I know it’s hard to believe this because it’s counter-intuitive, but the hard data say it’s true:
Raising the stakes on test scores does not improve student achievement. NCLB has failed to raise national test scores, even though, all across the country, schools are putting enormous effort into raising test scores, much more than ever before.
If we want “data-driven” policies, we can’t ignore the evidence.
At best, NCLB has raised scores on the particular tests used in each state to determine AYP in that state, but not on another test covering the same subject matter. Students are not learning more, they’re just getting more savvy about one test.
Schools should help kids learn that reading is wonderful, that math is fascinating, and that both have many uses, as do the subjects that standardized tests don’t cover.
Somerville teachers do that every day. Let’s not push them off stride with pressure for higher scores.
loading...
You should visit the schools. There is a real disconnect between some of the elementary schools and the high school, which is the subject of intense planning. Yet, except for some of the elementary schools, there is remarkably little focus on test scores, and far, far more on attendance, on motivating curiosity, and on building collaborative, responsible, inquiry skills. The obsession with tests is limited to schools where those tests are badly managed, and where poor scores – usually reflecting ELL and SPED conditions – have drawn too much attention from administrators, largely in response to parental concerns.
In fact, Somerville schools are exceptionally independent of most of the fears you have so well documented that are undermining national education policy. In part that’s possible because most kids “pass” MCAS with some regularity, and so the pressure – even of “low performing” – is toward higher goals. Yet, particularly in the high school, the focus is on soft skills – responsibility, teamwork, creativity, inquiry, work across cultures, and the like. Those are not “soft” – they can be measured, documented, and used to structure a wide range of learning activities, by kids as well as by teachers, parents, and others. And that is precisely what is happening in most classrooms, and in very, very many after-school projects and activities.
When the Director of Race to the Top was in town just about a year ago, to present their thinking on Assessment at a series of workshop events, Linda Darling-Hammond, of Stanford and formerly of Columbia, challenged the Department to explore portfolios. She cited several models with which we have actually worked, and explained, quite nicely but compellingly, that portfolios are far better evidence of student achievement – in ALL fields – than tests, and that portfolios can be assessed, evaluated, documented, and shared at least as well and as concisely as a set of scores to tests that have only tangential relevance to curriculum.
The response from the Department was “if we don’t spend our budget on tests, what will we spend it on.”
In contrast the Somerville High School Council authorized electronic portfolios, the pilots of which will be available in December, and the long term impact will eventually show – in quite clear terms – student growth in precisely those areas colleges and employers most seek, and, not coincidentally, precisely where SHS teachers and students excel. Rely on the fact that we have plenty of data that exceed the quality and quantity of test scores, and which will, in the next few years, support higher rates of college to more competitive colleges with higher levels of financial aid for all Somerville students.
loading...
See today’s AP article in the Globe State throws out MCAS results at Worcester school.
loading...