T or F? The standardized test conundrum



    pencil with side titlesThe exam starts nowStandard operating procedureTough testing in TexasThe public voiceSeeking alternatives














10 years from now, many of our best teachers and parents will be in private schools.




Which is a good alternative to standardized tests?

aStudent portfolios

bStandardized assignments

cBetter validation of tests

dMore reliance on grades

eMore (gasp!) computers in testing

Supporters say standardized tests are the only way to account for the enormous sums spent on public education, that tests hold to the grindstone the noses of teachers, administrators and students while helping instill a climate of competition that benefits all sides.

part of a standardized test, the example, showing the correct way to mark the circles on a test"If you remove standardized tests, what is the alternative?" asks education researcher Mike Antonucci. "The difficulty you run into is that we have standardized tests because they are standardized, and you can give the same test in California and Maine, get some idea of how they are learning particular subjects."

On the other hand is a group of educators and education researchers who think that more standardized testing is the road to perdition, transforming education into a cookie-cutter technology based on manufacturing principles rather than the goal of fostering and fulfilling curiosity. "It's a horrible tragedy," says Linda McNeil of Rice University. "The terrible part is that 10 years from now, when people say, 'We've got to reform our schools,' but by then, many of our best teachers and parents will not be in public school. We see them walking across the street to private schools already."

Testing the alternatives
Any effort to substitute better evaluations for standardized tests faces considerable opposition, inertia, and the lack of a simple alternative. The Kentucky Instructional Results Information System was designed to measure performance, not just the ability to take tests. KIRIS produced radical, rapid improvements in achievement. Unfortunately, the changes were not reflected on other standardized tests, leading a Rand Corp. assessment to conclude that KIRIS had evinced score inflation -- a major failing of non-standardized evaluation.

Some observers think standardized tests may be a good part of a more comprehensive evaluation system. Says Rand's Brian Stecher, "Commercial standardized tests are not a bad indication of the broad range of things that are relevant to education, but [when used in high-stakes systems], their flaws are exacerbated." Stecher suggests auditing standardized tests more closely by having "a sample of kids in sample classes take alternative measures that the teacher has not seen before." This would, he says, show whether the standardized tests in use are really measuring what we think we are measuring, i.e. the concepts we want students to learn. "The key question," Stecher says, "should be how well do standardized tests portray the broad range of outcomes we hope school will effect, and the broad range of skills we hope kids will master."

Instead of making tests the Holy Grail --

aA type of grain used for the king's ale,

bAnother term for "will o' the wisp,

cA holy pursuit or

dA random set of letters

-- Stecher says, "I believe we need to put more trust in things like grades and classroom work." For example, teachers could use a few standard assignments that would be common to all classes. "You'd let teachers grade and incorporate it into their instruction, but it would provide more directly comparable information" between schools.

A related idea, promulgated by Indiana's David Gilman and others, uses student portfolios for evaluation. Comparing work from throughout the school year, Gilman says, gives a better idea of true achievement.

illustration of many computers crowded together and with standardized tests on their screensIronically, one answer to the glut of computerized testing may be more computers, in a strategy called "adaptive testing." When a student takes a test, Stecher explains, "there are lots of questions that are too easy, and lots that are too hard, and that wastes time.î Why bother asking for the mass of the Earth in picograms when the student can't convert grams into kilograms? Adaptive testing would put banks of questions on the computer, which would branch to the student's level of knowledge. "It only asks more questions whose difficulty is close to the studentís own ability," and the result is to shorten testing time considerably without sacrificing the benefits of testing.

We've played our hand. Now it's your turn. Get out your standardized pencil and tell us.

Standardized tests are:

aA blessing in disguise,

bA foot in the door,

cA thorn in the side, or

dA burr under the saddle.



    The Why Files     There are 1 2 3 4 5 pages in this feature.
Bibliography | Credits | Feedback | Search