. . . and one test to rule them all . . .

October 18, 2012

The fundamental problem with standardized testing is that kids don’t come “standard.”  A teacher friend sent me this cartoon, summarizing the problem nicely:

One test fits all cartoon

If we use tests to cheat teachers out of a job, and to cheat students out of their futures, should we be surprised when students and teachers cheat on the tests?  Is that a mark of their innate sanity, if they do?

I wonder who the cartoonist was, and what other gems may be lurking out there by the same person?

Tip of the old scrub brush to Daniel Rhee.


Test “priming”: Malcolm Gladwell on how to push test results, and why tests might not work

July 31, 2012

Who is the interviewer, Allan Gregg?

From the YouTube site:

Malcolm Gladwell in an interview about Blink explains priming, and re-states some of the examples of priming from Blink with CC (closed captions)

Here’s a longer excerpt of the interview; from TVO (TVOntario)?

Discussion:  Gladwell appears to confirm, for testing results, the old aphorism attributed to Henry Ford:  “If you think you can, or if you think you can’t, you’re right.”  Gladwell seems to be saying that the student’s view of his or her abilities at the moment the test starts rules in a significant way how the student performs — worse, for teachers, it’s the student’s unconscious view of his or her abilities.  As a final shot in class, I have often had students predict their performance on state tests.   I have them write what they think they will scores.  Then I ask them to predict what they would have scored, had they applied themselves seriously to study of history — and of course, almost always the students have a fit of honesty and predict their scores would have been higher.  Then I ask them to pretend they had studied, and cross out the lower predicted score and replace it with the higher predicted score.  At the schools where I’ve taught, we do not administer the tests to our own students, and such exercises are prohibited on the day of the test.  Too bad, you think?

Another exercise I’ve found useful for boosting scores is to give the students one class period, just over an hour, to take the entire day-long TAKS social studies test, in the on-line version offered by the Texas Education Agency.  Originally I wanted students to get scared about what they didn’t know, and to get attuned to the questions they had no clue about so they’d pick it up in class.  What I discovered was that, in an hour, clearly with the pressure off (we weren’t taking it all that seriously, after all, allowing just an hour), students perform better than they expected.  So I ask them to pass a judgment on how difficult the test is, and what they should be scoring — almost unanimously they say they find the test not too difficult on the whole, and definitely conquerable by them.

What else could we do with students, if we knew how to prime them for tests, or for writing papers, or for any other piece of performance on which they would be graded?

With one exception, my administrators in Dallas ISD have been wholly inuninterested in such ideas, and such results — there is no checkbox on the teacher evaluation form for using online learning tools to advance test scores, and administrators do not regard that as teaching.  The one exception was Dorothy Gomez, our principal for two years, who had what I regarded as a bad habit of getting on the intercom almost every morning to cheer on students for learning what they would be tested on.  My post-test surveys of students showed those pep talks had been taken to heart, and we got much better performance out of lower-performing groups and entire classes during Gomez’s tenure (she has since left the district).

Also, if psychological tricks can significantly affect test scores, surely that invalidates the idea that we can use any test score to evaluate teacher effectiveness, unless immediate testing results is all we want teachers to achieve.  Gladwell said in this clip:

To me that completely undermines this notion, this naive notion that many educators have that you can reduce someone’s intelligence to a score on a test.  You can’t.

More:


“Is our media learning?” Test scores require raises for public school teachers, don’t they?

September 10, 2011

Mike the Mad Biologist makes the case succinctly and clearly (teachers, observe his methods):

That shudder you felt was the Earth wobbling as an . . .

. . . education story actually covered U.S. students’ academic achievement during the last few decades accurately. I’ve made the point before that the claim of stagnating test scores for U.S. students is demonstrably false–in every demographic group, there has been a rise in achievement (and the minority-white achievement gap is closing to boot). Shockingly, in a Slate report on Steven Brill’s new book Class Warfare, Richard Rothstein sets up Brill with this:

The case they make for their cause by now enjoys the status of conventional wisdom. Student achievement has been stagnant or declining for decades, even as money poured into public schools to improve teacher salaries, pensions, and working conditions (reducing class sizes, or hiring aides to give teachers more free time). Teachers typically have abysmally low standards, especially for minorities and other disadvantaged students, who predictably fall to the level of their teachers’ expectations. Although teachers’ quality can be estimated by the annual growth of their students’ scores on standardized tests of basic math and reading skills, teachers have not been held accountable for performance. Instead, they get lifetime job security even if students don’t learn. Brill observes a union-protected teacher in a Harlem public school bellowing “how many days in a week?,” caring little that students pay him no heed and wrestle on the floor instead.

Protecting this incompetence are teacher unions, whose contracts prevent principals from firing inadequate (and worse) teachers. The contracts also permit senior teachers to choose their schools, which further undermines principals’ authority. Union negotiations have produced perpetually rising salaries, guaranteed even to teachers who sleep through their careers. Breaking unions’ grip on public education is “the civil rights issue of this generation,” and some hard-working, idealistic Ivy Leaguers and their allies have shown how.

And then knocks him down with:

Central to the reformers’ argument is the claim that radical change is essential because student achievement (especially for minority and disadvantaged children) has been flat or declining for decades. This is, however, false. The only consistent data on student achievement come from a federal sample, the National Assessment of Educational Progress. Though you would never know it from the state of public alarm about education, the numbers show that regular public school performance has skyrocketed in the last two decades to the point that, for example, black elementary school students now have better math skills than whites had only 20 years ago. (There has also been progress for middle schoolers, and in reading; and less, but not insubstantial, progress for high schoolers.) The reason test score gaps have barely narrowed is that white students have also improved, at least at the elementary and middle school levels. The causes of these truly spectacular gains are unknown, but they are probably inconsistent with the idea that typical inner-city teachers are content to watch students wrestle on the classroom floor instead of learning.

The question we need to ask is “Is our media learning?” (to steal a phrase from Little Lord Pontchartrain).

Maybe they are . . .

Brill, God bless him, proposed to shake up public schools in America a few weeks ago in a long article in the Weekend Wall Street Journal.  His solution?  Make AFT local leader Randi Weingarten superintendent of New York’s public schools.

Actually, his story was much better than his advocacy.  But I hope to get more commentary on that proposal, and this continuing War on Education and War on Americans, soon.


A Seussian fable, about how tests ruin school reform

April 7, 2011

Look, you really need to go to TeacherSabrina’s blog, Failing Schools, where she first posted this, and take a look at many of her posts.

But I can’t resist putting the video here, because I know a few old-timers would be confused by the link, and some people will think they don’t have the time to make two clicks instead of one.

So, here, in its Seussian glory and demand for Flash Animation, is TeacherSabrina’s story of D.C.’s late Queen of the Schools, Michelle Rhee, and her desire to get D.C.’s kids to score well on an increasing battery of tests  [Got a good joke about assault and battery we can insert here?]:  “Rhee the Reformer:  A Cautionary Tale.”

Tip of the old scrub brush to Accountable Talk.


NY story: More money = better schools, better tests

June 24, 2008

Headlines across New York this morning shout about improved test scores, especially in reading and math, almost across the board. Scores are up in the “troubled” schools of New York City (and in the less-mentioned untroubled schools), scores are up in Buffalo. The news is so universally good that some are worried about statistical goofs, or cheating.

And while most economists with the possible exception of Milton Friedman would think it’s not news, some people point out that scores are up in poorer districts that got more money for educational programs.

At a news conference in Albany, the state education commissioner, Richard P. Mills, called the results “encouraging and exciting,” saying they were evidence that the state’s emphasis on giving more money to poorer school districts and focusing on high standards was successful. “The schools have delivered,” he said. (New York Times)

The War on Education continues unabated, however. The headline in the New York Daily News: “State math and reading exam scores released; critics question improvements.”

A beleaguered parent commented at the Daily News site:

our children are forced to do homework over weekends, Christmas vacation, winter break, spring break, etc. to prepare for these tests – their scores are up because they’ve worked hard all year!

If we’re wise, we’ll applaud the students and teachers, and we’ll wait for a lot more analysis. NYC Educator? JD2718? Is this good news for teachers in New York? Good news from teachers?

Update, June 26: NYC Educator takes note, “Test Scores Explode Statewide.”  JD2718, “Integrated Algebra Conversion Chart, Later Today.” Also:

Other notes and resources:


The best way to study for a test

December 7, 2007

Cognitive Daily dances through the research with alacrity, pointing to some research-approved methods for studying to do well on tests.

The best way? Greta and Dave Munger, the authors at Cognitive Daily, show the results that say students should study, take a month off, then study again. Cramming the night before has extremely limited benefits.

Can you apply that in class? Will your students listen to you?

The No Child Left Behind Act makes rumblings about using only research-proven methods in the classroom. If anyone ever enforces that clause, this post at Cognitive Daily better be your most visited site on the web. (They have other links, too. See “The Science of Cramming.”)

And, maybe we ought to stay up on the issue — the Mungers posted that information way back in August . . .


%d bloggers like this: