Multiple Measures More Rigorous Than Standardized Testing Alone

Some high profile folks have expressed dismay at the possibility that the Federal government may be considering using multiple measures of student performance, rather than the current reliance on standardized test scores alone. Eric Hanushek of the Hoover Institution recently shared his concerns in a letter to House Ed Committee chairman George Miller.

With all due respect to Mr. Hanushek, he might need to check across the campus at Stanford and talk to some of his colleagues such as Linda Darling-Hammond or Pam Grossman on whether it is truly in the best interests of our country to continue focusing on basic skills using only the results of standardized testing as key indicators of student achievement, and indirectly of teacher quality.

The current testing system, flaws and all, leaves the work of the majority of teachers totally unexamined. Standardized testing is not nearly rigorous enough or comprehensive enough to tell us what we really need to know to improve teaching and learning in our public schools. I have a son who has been able to pass the written portion of the state driving exam since he was eight, but still can't drive well enough to get the license. The reverse is also true in many cases: students who can perform well, but can't pass certain types of tests. I find out whether my students know how to write a researched paper, by having them write one (or several). More important, I check their progress along the way (formative assessments) and provide additional instruction or support as the results indicate.

There are many effective teachers out here whose work is not reflected in the test scores, and who regularly use differentiated instruction and perfomance-based assessments. There are entire teams of teachers, and schools which do so. In short, our myopic focus on standardized testing keeps us from getting a truer picture of what our students and our schools may actually be doing (good and bad). It is a well-documented fact that standardized tests are notoriously poor indicators or predictors of the academic performance of students of color, particularly African American students.

Multiple measures would include a combination performance assessments and standarized test scores. While performance assessments in education are sometimes decried supposedly for lack of reliability, that issue has been addressed and resolved successfully for some time in a number of fields, some at large-scale. However, these more comprehensive and rigorous measurements are more expensive and not as profitable to those benefiting from the current testing system.

Like any reform, NCLB included, the use of multiple measures should be approached thoughtfully and implemented carefully--both of which are more likely to happen if highly accomplished teachers are included throughout the process.

Comments

[Nancy Flanagan]

I have only one response to your blog, Renee: to point you and repeat "What She Said." In fact, we seem to be on parallel tracks on the "multiple measures" thing:

http://tinyurl.com/2adl7y

[Bob]

I agree that many people interpret some research findings to mean, "It is a well-documented fact that standardized tests are notoriously poor indicators or predictors of the academic performance of students of color, particularly African American students."

I suspect that you also know that others, who work from primary sources starting with the earliest standardized tests understand that they predict within identified measurement errors what students most likely will accomplish in schools in the future. Yes, some ethnic and cultural subgroups continue to score above and below 67 percent of other students.

Neither view necessarily predicts future academic performance of any individual with 100 percent accuracy. Correct?

[TeachMoore]

Actually, Bob, it was from my testing expert sources (who do work with primary sources and have been involved in the development of many of these standardized tests including ACT, NAEP), that I got that fact and stand by it: standardized tests are NOTORIOUSLY poor indicators or predictors of the academic performance of students of color, particularly African American students. Not that they don't try to do the best job they can with the tools they have. I have been an item writer and a test item reviewer for ACT for many years.

My point, however, is that a standardized test alone cannot be a truly rigorous, accurate measure of student achievement or progress. A combination of measures, thoughtfully analyzed by a highly accomplished classroom teacher can.

[Bob]

Congrats for contributing test items to pools for constructing tests. I hope your contributions help shape distributions of test scores. I agree, Renee, that many have for several decades interpreted standardized tests in ways you mention and have added a sizable body of empirical literature to underwrite such conclusions. Others have considered standardized test results differently and as authoritatively since the 1920s. I look forward to further research clarifications of factors that influence standardized test score distributions.