Alfie Kohn has written many books critical of competition and ranking in schools. This article appeared in the New York Times.
For a generation now, school reform has meant top-down mandates for what students must be taught, enforced by high-stakes standardized tests and justified by macho rhetoric — “rigor,” “raising the bar,” “tougher standards.”
Here’s a thought experiment. Suppose that next year virtually every student passed the tests. What would the reaction be from politicians, businesspeople, the media? Would these people shake their heads in admiration and say, “Damn, those teachers must be good!”?
Of course not. Such remarkable success would be cited as evidence that the tests were too easy. In the real world, when scores have improved sharply, this has indeed been the reaction. For example, when results on New York’s math exam rose in 2009, the chancellor of the state’s Board of Regents said, “What today’s scores tell me is not that we should be celebrating,” but instead “that New York State needs to raise its standards.”
The inescapable, and deeply disturbing, implication is that “high standards” really means “standards that all students will never be able to meet.” If everyone did meet them, the standards would just be ratcheted up again — as high as necessary to ensure that some students failed.
The standards-and-accountability movement is not about leaving no child behind. To the contrary, it is an elaborate sorting device, intended to separate wheat from chaff. The fact that students of color, students from low-income families and students whose first language isn’t English are disproportionately defined as chaff makes the whole enterprise even more insidious.
But my little thought experiment uncovers a truth that extends well beyond what has been done to our schools in the name of “raising the bar.” We have been taught to respond with suspicion whenever all members of any group are successful. That’s true even when we have no reason to believe that corners have been cut. In America, excellence is regarded as a scarce commodity. Success doesn’t count unless it is attained by only a few.
One way to ensure this outcome is to evaluate people (or schools, or companies, or countries) relative to one another. That way, even if everyone has done quite well, or improved over time, half will always fall below the median — and look like failures.
Kohn quite rightly concludes that the nature of the standards-and-accountability regime of federal policy (No Child Left Behind, Race to the Top, Every Student Succeeds Act) requires that most children are left behind, most children will never reach the top, and most children will not succeed. The reliance on standardized testing, normed on a bell curve, guarantees that outcome.
Please note that, as the title indicates, Kohn is not just talking about standardized tests. He’s talking about any form of ranking, including grades.
Very much so, Dienne! Thanks for pointing that out.
Why is it that the only letter grade in the grade scale, and it is a limited scale, “F” actually is a word with meaning?
Let’s see, a student does his/her very best and she gets an “F”. A failure he is. What is totally wrong with that picture?
Kohn is spot on — especially given the terrifying increase in teenage mental health hospital admissions and visits. What was beyond sad were the NYTimes reader comments to Kohn’s piece, pushing for homework and advancement. Why is it so hard to understand what the research says about homework as well as just listening to common sense: let kids use those early years to develop a sense of self, develop confidence, find the strength to feel failure failure, know it’s ok and keep moving on, form relationships, learn to fight and forgive. These are the building blocks of excellence in future. academics.
the terrifying increase in teenage mental health hospital admissions and visits
Thank you, thank you, thank you for raising this issue, Lisa. This is one of the truly horrific but rarely mentioned consequences of Ed Deform. The Deformers lack any motivations except money and power, so they don’t understand that other things motivate people, and they ignore the voluminous research showing that extrinsic punishments and rewards are actually demotivating for cognitive tasks.
Bell curve logic: we need bell curve tests to identify the geniuses among us, so we can make the best use of their special gifts. But if these kids are, in fact, geniuses, why do we need bell curve tests to identify them–isn’t the point of being a genius that one’s special abilities are obviously superior to those of the rest of us? Mozart, Bach, Newton…we found them (or they found us) without them being identified by a bell curve-type test. And for everyone else, why deny society the special talents that the 99% possess but are not valued by the folks giving the tests? God knows there are many, many people whose talents, though not especially important to the Koch brothers, are valuable to many other people. Makes one think that the real function of bell curved-type tests is to confirm and authenticate the standards of value of those in power, and not to support the full development of all children, be they Mozarts or tinkerers in garages or Martin Luther King Jr or the kids who figure out solutions to all sorts of (important) problems to which Common Core Standards pay little attention.
Re-read The Big Test, by Nicholas Lemann, to be reminded of the troubling origins of our famous bell curved tests, and the myths that have been concocted to sell them.
Scrap bell curved tests, and the cheap, class-based, educational system that calls sorting kids “education,” and creates instead an education system that invites children to create themselves with the help of teachers who know how to guide them, if given genuine support. Our failing society, which bell curved tests have helped to sustain, will not be improved by the system that generated these failings in the first place.
\We all know this truth, but we also know how much money and power stand behind the preservation of a system that has run its course. Do we really need bell curve winners to figure out how to corrupt our elections so the bell curve winners on the Supreme Court can explain why Madison and Hamilton wrote a constitution that proclaims such corruption as the essence of liberty?
There are good bell curve winners and bad bell curve winners.,..obviously we need something more than a bell curved test to distinguish between them. (Just think of all the bell curve winners who have supported, and created, Evil. I think Satan’s SATs were probably pretty damned good.) Time to admit that we need an education system that does not tell most kids that they have few if any important talents. (How stupid is that?)
Well said, Steve!!!
The Big Test is outstanding. I join you in recommending it highly. https://www.amazon.com/Big-Test-History-American-Meritocracy/dp/0374527512
From the post:
“If everyone did meet them, the standards would just be ratcheted up again — as high as necessary to ensure that some students failed. . . . The standards-and-accountability movement is not about leaving no child behind.”
And from Steve:
“Re-read The Big Test, by Nicholas Lemann, to be reminded of the troubling origins of our famous bell curved tests, and the myths that have been concocted to sell them.”
Mark Garrison in “A Measure of Failure” describes the political history through a political sociological lens of standardized testing and the inherent contradictions in attempting to achieve equality of results when the society surrounding schooling is inherently unequal. He shows how equity of opportunity through standards and testing malpractice regime does not and cannot achieve the goals to which the proponents claim to adhere.
It would behoove all here to read his spot on analysis.
You nailed it! And it is detrimental to so many children.
From “A Measure of Failure”:
“The representational status of our current assessment tools is revealed in their fixation on social value–the linking of individual place to social slot in a hierarchical and increasingly unequal social system. Like the rigid feudal society that predated them, testing techniques are fixated and validated on their link to unequal social structure while employing procedural equality through standardization. They were able to buy time for liberal political arrangements by transforming the process of marking social value to one that is “open” and “equitable” and undertaken and arbitrated by “neutral” authorities, while in fact structuring inequality on a historically unprecedented scale. Unlike fixed-class or caste systems, everyone had the right to be marked worthy or unworthy; the individual’s mental attributes were given as a natural (unbiased by social hierarchies) means by which to assign social value.
This contradiction is evident in the project of “closing the achievement gap.” The policy of closing “the achievement gap” is a legacy of the theory of natural distinction, although current policy works the theory backward. While academic achievement and ability were given as a natural basis for social distinction, today’s project seeks to remedy the overgrowth of social inequality by closing gaps in academic performance, that is, minimizing natural differences as evident in academic achievement.
But this strategy fails because social difference does not have its origin in natural difference for the simple fact that human beings are one group, not a collection of naturally occurring subgroups or “races.” Variation among individuals is, in fact, a within-group phenomenon (see Shiff and Lewontin 1986). Social differences are social in origin, and have no basis in “natural” distinctions of intelligence or even strength as social value distorts history by abstracting individuals from their social reality.
The achievement gap’s irrationality is a legacy of this theory. ATTEMPTING TO EQUALIZE SCHOOL OUTCOMES VIA MEASURES THAT PRESUME TO DIFFERENTIATE IS NOTHING IF NOT IRRATIONAL. . . . ” [
Oops–[MY EMPHASIS]
Ranking via testing is a tool to keep inequality in place. It’s a powerful weapon for mass management and discipline.
Exactly.
YES! I used to spend a lot of time (aka wasting my life away) commenting on Orlando Sentinel articles back when they had comments on their website (as opposed to Facebook or the “powered by Facebook” comments thing).
What Kohn described happened EVERY TIME there was a report about rising or falling test scores.
If they went up: “They lowered the standards.”
If they went down: “This is why we need reform.”
Regardless of what one does about the grading system, K12 goals need to be changed. The words “success” [/failure], “high-achieving” [low-achieving] and their corollary, “achievement gap,” need to be removed from the vocabulary of missions and reports. If they ever had any value, they have been tainted by association w/ the plague of annual stdzd test scores. Those words imply precise measurements exist for an ever-evolving, blessedly indefinable process. They bizarrely frame learning as a tournament sport, a zero-sum game, a bell-curve. K12 standards should be simply the baseline “be-able-to’s” required to pass/ graduate, period—the minimum tools needed for entry-level jobs &/or further study—and teachers will decide if they have been met.
Perhaps even get rid of “quality” [/adequate] as in “equal access to a ___ education.” I’d prefer “equal access to educational opportunity,” which points to the civil-rights function of the fed DofEd. Their other main function, gathering comparative regional stats, is already done by sampling w/NAEP, & can be augmented by diagnostic tests every 3-4 yrs as done prior to 2001.
RE: grades… For some reason, many commenters at the NYT were thinking college, even tho Kohn noted the problems w/NCLB policies, so probably had his mind on K12. Many inputs from profs complaining of admin pressure to inflate grades.
I see their issue as very similar to K12 grading problems since late ‘70’s: how do you grade all these “new” kids: no more tracking; you’ve now got to deal w/ good-intelligence kids hampered by diagnosed LD’s (they used to be in the slow track); how about the ESL’s you used to let sink or swim & now have to consider their intellectual capability; are your poorest students getting free supplies/ food/ clothing/ msw-guidance support, because poverty is a known factor in their ability to perform.
At the college level, older profs now deal w/many kids who were not considered college material when they were in college or started as profs: matriculation rates are 20% higher than in the ‘70’s. Many of the newcomers bring their LD/ ESL/ poverty [1st in family to go to college] issues with them. Grading is probably just part of their problem: they’re trying to figure out how to adjust their expectations, and their teaching methods.
The point of education should not be to stack rank kids according to some universal standard but, rather, to provide paths via which differing kids can realize their potentials. The concepts of standardization and relative ranking are wrong from the start.
The challenge is for us to envision a system that makes this possible–one that enables each child to find that path that works for him or her. A system predicated on rejecting an enormous number of children as misfits is just evil.
We don’t even have to imagine, just check out some of the successful Euro systems like Finland’s or Poland’s. The structure provides common education through age 16, after which there are multi-level branches in two directions: “pure” academics & applied sciences. The latter goes from job-ready voc cert level through advanced engrg/ med/ pharm etc degrees. You can move laterally between branches. Education is funded publicly through the highest level (including for any age adult in Finland). Goals here clearly are to educate everyone & help them find best path for themselves, not winnow out losers.
Yes. But I would like to see more fundamental change. For example, I would like to see grades for whole courses replaced by a great many certificates of achievement gained by production, to the satisfaction of teacher committees, of evidence of satisfactory attainment, such as portfolio work, successful completion of a role in a play, creation of a website using HTML, completion of a reading list, etc.
But yes, these are much saner systems!!!
I am curious about your use of the word “universal” here. Does it mean that you think stack ranking based on local standards, that is teacher assigned grades and the resulting gpa and class rank, is not “wrong from the start”?
Good point, TE. My wording was sloppy. No, of course not. As I mentioned above, I support replacing grades with demonstrations of accomplishments that earn certificates of achievement, conferred by local teacher panels, that accumulate, creating profiles appropriate to differing students on differing paths.
I would want the emphasis, in these, to be on products requiring knowledge and skill–an academic paper illustrating command of documentation style and format, a major auto repair, a website constructed using HTML and cascading stylesheets, an edited video, a completed course of reading in a particular subject area, etc.
I’ve seen too many kids with enormous gifts that were not recognized and built upon by our existing system. I’m thinking, for example, of a particular young lady who was in one of my classes last year. She wasn’t going to be a mathematician or a Milton scholar. Her highest grade in any class was a C. She had not passed her state exams in either math or English and was not likely, as a result, to graduate from high school. But she had many, many friends; a joyous spirit; a wicked sense of humor; a gift for hairstyling and makeup; enormous energy; and she was a talented gymnast. Our system was failing her because it did not recognize and build upon her particular genius.
A child can come into our K-12 system with perfect pitch and never have that recognized by anyone.
It does seem to me that folks here are concerned about prostate cancer in patients that are having a heart attack. Some students are directly impacted by staking and ranking by standardized tests, but ALL students are directly impacted by staking and ranking by teacher assigned grades.
If you are in Florida, there is no worry under state law about high school graduation for your student due to low standardized test score (see http://www.leg.state.fl.us/Statutes/index.cfm?App_mode=Display_Statute&Search_String=&URL=1000-1099/1003/Sections/1003.4282.html) . There is, however, a HUGE risk that she will not graduate because of a teacher ranking in any of the 24 credits that are required for graduation or if she has failed to achieve a 2.0 gpa.
A closer reading shows there are three ways a student might not graduate from high school due to stacking and ranking: 1) not getting a passing grade in a required course, 2) not having at least a 2.0 gpa, or 3) not getting a passing score on the grade 10 reading/ELA standardized exam.
Stacked ranking is an employee evaluation method that slots a certain percentage of employees into each of several levels of performance. Grading on a curve is stack ranking. “Teacher assigned grades and the resulting gpa and class rank” are not stack ranking—unless of course all teachers are mandated to grade on a specified curve [unlikely—I hope].
State standardized tests however are graded on a curve. According to this article, a high curve is set to compensate for the inevitable mismatch between content covered on the test vs taught in the classroom. https://athensoracle.com/1993/opinions/standardized-testing-isnt-a-fair-way-of-evaluating-students/ Obviously this does not magically convert the stdzd test score into an accurate measure, it just points up what a poor measure it is.
Then there’s the cut scores established annually by the state, after stdzd test scores are in. Not a curve, but clearly a huge fudge factor that can be used to show everyone’s above-average Lake Woebegone style, or [more likely] the sky is falling our kids can’t read. See https://www.washingtonpost.com/news/answer-sheet/wp/2014/04/29/the-scary-way-common-core-test-cut-scores-are-selected/?utm_term=.4b24c7b8517c
And of course you have the frequent changing & shuffling of stdzd test vendors and formats, making yr to yr comparisons false.
There’s undoubtedly inconsistency between classrooms, schools, districts and regions when teachers do the grading, & it has nothing to do with stack ranking. Administrators can make some corrections at the school level. NEAP, plus a couple of std diagnostic tests (2 or 3 total K-12) can tease out differences and help compare regions on a more accurate footing. State stdzd tests on the other hand are a poor measure of anything, swallow months of teaching time, and are a tax-funded boondoggle for sw/hw/testing corps—all to establish a seemingly superior (but faux) “consistency.”
Bethree5,
Of course teacher grades, high school gpa, and class rank are stacking and ranking.
In Florida the cut scores are not determined after the exam is given. The ELA cut scores there were set in 2015 and will only change if there is a major change in the exam. See http://www.fldoe.org/core/fileparse.php/5663/urlt/V3FSA1516TechRpt.pdf
The bell curve needs to be expunged from education. 99% of the population doesn’t know what it is, anyways.
Agreed. It’s extremely useful for some purposes, such as in measuring outcomes of industrial manufacturing processes, but in education it does a LOT of damage.
One step toward getting rid of grades: Mastery Transcript Consortium at http://mastery.org. We did a podcast with the founder Scott Looney. (https://modernlearners.com/mastery-transcripts-replace-grades/) In his words, there is no evidence that grades promote deeper learning. Just deeper schooling.
As a School Librarian in the Buffalo Public Schools I worked with the reading and classroom teachers in one of the elementary building encouraging the children to read a wide variety of books beyond those assigned as school work. They became excited when they discovered The Magic Tree House and Julie B Jones series as well as other high interest titles. I felt proud when their test scores went up, believing this was a result of us working as a team and motivating our students to improve their reading skills.
What a disappointment to discover that all our hard work resulted in tougher tests which were designed for failure (beyond their reading level where only the top thirty percent can “pass” with an acceptable score). No body believed that the results were a valid reflection of improvement – no, it must be that the tests were too easy.
And that was before the introduction of Common Core which only made things worse.
Junie B Jones