Testing... Testing...
If you believe school reform is urgent (and you should), then you should be in an absolute lather over the quality of our standardized tests. Here's why:
The tests are allowing us to stay in a state of permanent emergency response. Despite all our talk of high expectations, we've geared the system to minimum expectations for many of our students. Lousy tests have become de facto standards in too many places. As the pressures to make Adequate Yearly Progress build so will the pressure to narrow schools' vision. With many of the tests we have, it will be hard to tell the difference between the schools that stay in triage and those that lift their students to world-class standards. So much for transparency.
The tests are becoming the measure of everything, not just schools or Students. Every reform, every innovation, every old or new practice seems to rise or fall on the results of state tests. We make sweeping judgments about what works on the basis of tests, and we often use anemic (though "significant") gains in scores to proclaim one reform better than another.
Take the on-going debate about class size, for example. Research on the benefits of small class sizes is mixed. But Nancy Flanagan offers a bracing caution: "When our only measures of student success are memorized material, spit back on a bubble-in test, then a class of 45 listening to a teacher's lecture may be indistinguishable from a class of 25 listening to the same lecture."
There may be real problems with efforts to reduce class sizes. Where, for example, do we find all the extra teachers we need to make those efforts work? (If only all those class size skeptics who would have us fire scads of teachers in struggling schools would admit that their own favorite reform suffers from the same problem.) But Flanagan's point about test scores should give us pause.
In so many ways, we set our compass by those scores. They could dictate our direction for years to come. It would be far too easy to tolerate practices that improve test scores but not much else. Worse, we're tempted to enshrine those practices in policy.
I don't mean to knock standardized tests or the people who use them. We use them. We can't simply trust our intuition about what our students might be learning. We do need valid, reliable, and--yes--standard data about how our students are doing, or all the finest sentiments about what our students should learn won't do us any good at all.
But our current tests just aren't doing the job, and that fact has huge implications for everything else we do. If we're truly honest about the urgency of school reform, we should sound every available alarm bell about the quality of our tests.
SIGN UP
Visionaries
Click here to browse dozens of Public School Insights interviews with extraordinary education advocates, including:
- 2013 Digital Principal Ryan Imbriale
- Best Selling Author Dan Ariely
- Family Engagement Expert Dr. Maria C. Paredes
The views expressed in this website's interviews do not necessarily represent those of the Learning First Alliance or its members.
New Stories
Featured Story
Excellence is the Standard
At Pierce County High School in rural southeast Georgia, the graduation rate has gone up 31% in seven years. Teachers describe their collaboration as the unifying factor that drives the school’s improvement. Learn more...
School/District Characteristics
Hot Topics
Blog Roll
Members' Blogs
- Transforming Learning
- The EDifier
- School Board News Today
- Legal Clips
- Learning Forward’s PD Watch
- NAESP's Principals' Office
- NASSP's Principal's Policy Blog
- The Principal Difference
- ASCA Scene
- PDK Blog
- Always Something
- NSPRA: Social School Public Relations
- AACTE's President's Perspective
- AASA's The Leading Edge
- AASA Connects (formerly AASA's School Street)
- NEA Today
- Angles on Education
- Lily's Blackboard
- PTA's One Voice
- ISTE Connects
What Else We're Reading
- Advancing the Teaching Profession
- Edwize
- The Answer Sheet
- Edutopia's Blogs
- Politics K-12
- U.S. Department of Education Blog
- John Wilson Unleashed
- The Core Knowledge Blog
- This Week in Education
- Inside School Research
- Teacher Leadership Today
- On the Shoulders of Giants
- Teacher in a Strange Land
- Teach Moore
- The Tempered Radical
- The Educated Reporter
- Taking Note
- Character Education Partnership Blog
- Why I Teach
"We do need valid, reliable,
"We do need valid, reliable, and--yes--standard data about how our students are doing, or all the finest sentiments about what our students should learn won't do us any good at all."
You're right, which is why we can't do things like performance assessment and portfolios, which is what most of the anti-testing crowd is talking about now. They're not reliable, they're not valid, people do know how to do them well, and they'll set us way back in accountability. At least the "bubble-in" tests give us actual, reliable information.
How innnnnteresting. And
How innnnnteresting. And love how the first commenter in that link basically says that the only people who have opinions about education that matter are public school teachers. Odd that people sometimes think of homeschoolers as being arrogant when a simple perusal of the NEA website and a sampling of about 50 teacher blogs would tell you that they're not the only ones with this tendency.
Standardized tests are A measure, but probably not THE measure, of student success and learning. :)
" At least the "bubble-in"
" At least the "bubble-in" tests give us actual, reliable information."
Well, what information is not actual information? "Reliable" just means performance doesn't vary in random, unpredictable ways, which is true of some tests more than others, for some students more than others - assuming they care about their performance and do their best each year - which I know for a fact they don't, having administered those tests to several hundred students in the past decade or so.
We shouldn't necessarily lump in all "bubble tests" together - but many of them do not measure what they purport to measure, and the new reformers who get a little excited by the word "accountability" love to make all sorts of claims about those bubble tests. Their bias may be harder to detect to the casual observer, but it is every bit as real as the bias you fear in individual teachers or schools. Furthermore, many of the proposed uses for the tests are decidedly invalid. That is not just my opinion as a teacher, but also the consensus of the leading professional organizations in educational research and measurement. As an English teacher, I'll give you my main gripe, though there are many more. The tests my students are not adequately linked to the standards I'm supposed to use in my teaching. I have responsibilities in four areas of language arts, but reading is the only one of those that I would concede that the state tests might be measuring. So, right away, the tests fail to measure about three-fourths of my teaching domain. Then, they're not instructionally sensitive. Many of my students come to my class already possessing the skills to perform well on the tests, and all year long they practice those skills in a variety of settings, only one fraction of which falls under my control as a teacher.
So what does the test show? Once you factor in margins of error for any given sitting, you have an approximation of a biased and flawed measure of a seriously limited subset of a highly complex set of knowledge and skills influenced by an unmeasurable combination of factors.
But hey, it's actual information!
Oh Mrs. C--everyone who
Oh Mrs. C--everyone who speaks about education thinks their opinions are the only ones that matter. The real question is whose opinions have big money behind them....
Thanks for your very thoughtful review of the problems with current assessments, David. Do you think there's strong hope for an alternative systemthat captures what students should know and be able to do--and that can satisfy people's desire for accountability?
Post new comment