« Two Must-Read Studies on Teachers | Main | Peer-Assistance and -Review: The Toledo Numbers »

Performance-Based Assessment and Teachers


As I travel and talk to teachers, they consistently tell me that one of their biggest frustrations is the testing under the No Child Left Behind Act. Such testing is largely dominated by multiple-choice questions, and teachers feel under pressure to "teach to the test" or prep students for these kinds of questions.

As some of you know, in addition to covering teacher issues here at Education Week, I also track and write about the latest developments in student assessment. I just wrote a long story on researchers' ideas about how to improve assessment.

The germ of this idea came out of the notion that some tests might be worth teaching to, if they reflected the kind of rich activities and critical problem-solving that we want students to engage in and teachers to foster.

Although none of the examples I wrote about gets into the realm of "portfolios" or extended research projects—which have been shown to be somewhat unreliable and not comparable across states as an overall measure of student learning—they are all examples of extended performance-based tasks that require students to use critical problem-solving skills. And they are standardized, which means that they might be adopted for use in an accountability context.

Some will protest that tests should be used only for informational purposes, not for accountability. I understand those arguments, but as it's unlikely that test-based accountability is going away, I tried to tailor the story to what policymakers and researchers might be able to accomplish within the existing framework.

I hope you'll check it out and post your thoughts below.


The major drawback to testing for accountability is that such testing is not especially reliable for the purpose of accountability. Test results, regardless of the format give a snapshot of any one student's progress at a given moment. The results may beaffected by so many factors that it is difficult to pinpoint any one educational factor that accounts for anything.
The big hurdle is that tseting is output based. It does not offer a clear picture of input. The achievement of eleventh grade students on a standardized exam does not clearly reflect the teaching of eleventh garde teachers. These same students have had, at the least, twenty odd teachers in at least that many classrooms prior to entering eleventh grade. All of these teachers and grade levels have contributed to the student's achievement and level of learning. This is just one factor that is difficult to assess.
Other factors are just as difficult to categorize as influences. Socio-economic status can affect achievement. It can be both positive and negative, depending on the individual student. Standardized testing has even taken the step to separate achievement by ethnicity and skin color as if such cahacteristics are variables that somehow can be controled for, suggesting that if students' race could somehow be magically changed, their achievement could be improved.
Education is best served by investment in school quality and infrastructure, not wasteful spending on"achieve,ment tests" that really do not measure any such thing. Schools need desks, chairs, quality AV equipement, clean healthful grounds, textbooks, up to date technology, and teachers that are trusted professionals that have themselves been trained and tested

I find that authentic assessment is most useful and helps the learning process better than standardized assessments.The problem is that authorities need rewards on some evidence of performance that seems discriminative as historically proven. Therefore, authorities keep on insisting on standardized testing. It would be more realistic and professional to keep to authentic assessments only.

Using test scores to hold teachers accountable for educational progress presumes one thing that is not in evidence and overlooks many other things that are.

For statistical data from testing to mean anything about the teacher as a variable, every classroom must have the same cross sections of skills, abilities, disabilities, handicaps, and strengths. As we all know, class lists are subject to a multitude of variables that mean no teacher's class lists mirror any other teacher's lists, let alone all other teachers. Since those class lists contain such uncontrolled variables, you can't ascribe success or failure to the classroom teacher from test score data.

Furthermore, among those wide variations in student profiles are many things, over which teachers have no control, but that have significant influence on the students' prognosis for success in the class room. Again, there would be no validity in the statistics, since you can't isolate the variables. The results would tell you more about how student class lists are created than they would about how well the teacher in question actually taught.

Performance based assessments are critical weapons in the ELA teacher's arsenal. But, far more explosive are diagnostic assessments.
Diagnostic assessments are essential instructional tools for effective English-language Arts and reading teachers. However, many teachers resist using these tools because they can be time-consuming to administer, grade, record, and analyze. Some teachers avoid diagnostic assessments because these teachers exclusively focus on grade-level standards-based instruction or believe that remediation is (or was) the job of some other teacher. To be honest, some teachers resist diagnostic assessments because the data might induce them to differentiate instruction—a daunting task for any teacher. And some teachers resist diagnostic assessments because they fear that the data will be used by administrators to hold them accountable for individual student progress. Check out ten criteria for effective diagnostic ELA/reading assessments at http://penningtonpublishing.com/blog/reading/ten-criteria-for-effective-elareading-diagnostic-assessments/ and download free whole-class comprehensive consonant and vowel phonics assessments, three sight word assessments, a spelling-pattern assessment, a multi-level fluency assessment, six phonemic awareness assessments, a grammar assessment, and a mechanics assessment from the right column of this informative article.

Comments are now closed for this post.


Most Viewed on Education Week



Recent Comments