What is ED’s Stance on Using Testing Data in Teacher Evaluation?

A recent letter to the Department of Education from a teacher in Cincinnati contained a quote that really struck me: “It is not at all that I am afraid of what my test scores might reveal.  I am more concerned about what my student’s test scores will not reveal.”

The quote rings true of so many classrooms across the country, including my own. I teach students who have been removed from other institutions due to behavior, chronic absences or other issues that have prevented them from being successful in the traditional school setting. Each of my students has been identified as a potential dropout and each has a profound set of challenges that manifest in the classroom.

Marciano Gutierrez

Marciano Gutierrez is a 2012 Teaching Ambassador Fellow, on loan from Alta Vista High School in Mountain View, CA.

As a U.S. Department of Education Teaching Ambassador Fellow I have been able to engage with Secretary Duncan’s senior staff and have learned more about the Department’s stance on teacher evaluation. Like most teachers in the United States, Secretary Duncan strongly believes that a single test result does not adequately reflect the quality or complexity of excellent teaching.

At a speech to the National Council for Social Studies, Mr. Duncan stated, “Just to be 100 percent clear—evaluation should never be based only on test scores. That would be ridiculous. It should also include factors like principal observation or peer review, student work, parent feedback. It should be designed locally—and teachers should be at the table to help design it.” The Department’s work on educator evaluations has thus been to promote multiple measures to elicit a well-rounded perspective on one’s craft and to encourage districts and schools to primarily use these tools as a means for quality professional development. This thinking was also captured in a speech that the Secretary made to Baltimore County teachers this past fall.

As a teacher of students who historically struggle on standardized tests, I understand the concern about tying testing data- which is often influenced by factors outside of my control- to my performance. I am also sometimes frustrated by the quality of the multiple-choice assessments used to assess my students’ learning which are ultimately a reflection upon my practice. Despite these challenges, I do believe that there does need to be some measurement of student performance and growth. This information should be collected and analyzed so that we can continuously improve the learning experience for all students and to ensure that we hold ourselves to high standards and continuous improvement.

While the Department’s policy has been that measures of student growth and gain should be a ‘significant’ factor in teacher evaluations, the Secretary has said that, “we intentionally leave that undefined—because different states will have different approaches—and different confidence levels in their assessments.”

As a previous Teacher Fellow with the Hope Street Group, and in my current work with Race to the Top states, I have seen a variety of state-developed approaches and strategies that aim to meet this vision. I have come to realize that the strongest evaluation systems have been developed with robust teacher input at every stage of the process.  These evaluation systems, which are designed and improved with the practical insight of teachers, use test scores as only one of multiple measures of effectiveness, therefore allowing teachers of students like mine, to demonstrate quality teaching in ways that transcend test scores alone.

Marciano Gutierrez is a 2012 Teaching Ambassador Fellow, on loan from Alta Vista High School in Mountain View, Calif.


  1. Critic makes a good point, but it is NOT “punishment” for ED to tell states that it is a misuse of taxpayer dollars to have test scores be fify percent of a teacher evaluation.
    It would be good stewardship of federal funding.

  2. The use of standardized assessments in teacher evaluations is an inequitable practice. Additionally, the assessment is not a true reflection of a teacher’s ability. Standardized tests are cumulative and assess information covered in multiple grades. Some grade levels or content areas are not subjected to high stake tests; therefore the art, music, Kindergarten, etc. teacher is evaluated differently than the high school math or 5th grade classroom teacher. In the elementary grades, more than one standardized test is given- language arts, math, and sometimes science. In the high school, although in the same content area, teachers may be working with students taking various standardized assessments. What about special education teachers? I agree with raising the bar and ensuring our special education students receive access to rigorous curriculum standards; however, their ability to demonstrate skill acquisition will vary in comparison to their typical peers. Is it fair to use standardized scores to evaluate the special education teacher in the same manner as the general classroom teacher?

  3. Classroom teachers should not be assesse on a standardized test that does not include the objectives that the states have established. Teachers should be evaluated in several ways. One, teachers should be assessed on to what extent they are able to promote student learning in the classroom based on how the students view the teacher. We all know that when students like their teachers they learn. When teachers know the content they are teaching and care about their students, students learn. Secondly, teachers should be assessed on to what extent the students learn the objectives that have been established based on a criterion-referenced test that is developed by the school district or state. Now that the Common Core Standards are available, a test designed from those objectives should be used. Third, teachers should be allowed to assess themselves based on what they said they were going to do in the classroom. They know if they have managed their class in a positive manner that is conducive to student learning. In fact, they know when they are doing a good job, When they do, students learn. Fourth, teachers should be observed by their peers to determine to what extent they are doing a good job to promote student learning, and finally, this report should be written up and submitted to the principal, who will then observe the teacher to obtain the principal’s perspective of what is happening in the classroom.

    School districts should never used one test to assess the performance of teachers. Students are not machines, and teachers do not always have direct control over what students learn. Some students learn when the information is presented one time. Other students may need to be re-taught several times before they get the concept in their mind’s eye. So, those school boards who are trying to use one test to assess the performance of teachers should now realize that there must be more than one mechanism in place to evaluate teachers in the future.

    Dr. Cathine G. Gilchrist, Former Teacher

    • Thank you. Standardized tests do not measure student achievement or growth and therefore should not be used even as one of multiple measures to evaluate teachers. Teachers should be united on this to defend public education as an institution and all the children served by public schools.

    • Dear Dr. Gilchrist, Your mastery of the English language is incomplete. Please have a student read and correct what you write before you publish. Thank you

  4. What administrators (and that includes Duncan et al) and teachers must realize is that standardized tests come with a built-in (and guaranteed!) failure rate: the placement of the “proficient” cut-off point. In California, it is placed at the 50% point while in New York is at 40%.

    I have yet to see any discussion of this fact and its consequences. What do we do with those who don’t meet proficiency? Should they be forced to repeat the grade until they are proficient? (Do we have enough classrooms to do that?) In California, nothing is done since “promotion” to the next grade is based solely on classroom marks. This brings another issue: classroom marks do not correlate with test scores. Why? Because, through the magic of test design and statistics, the tests are not direct measures of academic achievement but rather the ability of the test taker in choosing a “correct” answer from several equally plausible choices.

    We need to have a conversation on the validity of standardized tests and why parents and/or teachers cannot question their contents. Enough of sorting and ranking without telling the public that their children are being graded “on the curve” (aka the Bell Curve, or, to the initiated, the Normalized Gaussian Distribution). Else Common Core will be a disaster.

  5. I feel the only way they should be included in the evaluations are on pre-tests and post tests. That way administration can see marked progress in the classroom. To include a statewide or nation wide standardized test without a teacher of any subject not knowing what would be questioned on that test and what students should be knowing and not knowing is unfair to the teacher especially when a non-tenured or tenured teacher’s job is on the line. If that is the case the Federal and state department of educations should make a test for every grade and explain their expectations when evaluating teacher performance in the classroom.

  6. I will be assuming a Trusteeship in an elementary district that feeds Marcino’s superb alternative-path high school. And the elementary district does not use any test metrics in their teacher evaluations or principal evaluations! Currently, the administration does not seem to compile the excellent state teacher rubrics into “easy for trustees to understand” scales. The evaluations go onto paper, into file folders, and forever disappear.
    I now realize, the Trustees need to first have evaluations for themselves and have the principals and teachers evaluate them! I think we need to work “top down”. Ourselves, district administration, principals – then the teachers with a very inclusive process. If we do that – I think the Trustees will gain the trust of teachers.
    CA is not a RTT state and my district has not taken the reform necessary to be a RTT-D $ contender.
    When I worked in industry, the US Department of Commerce was instrumental in offering excellent guidance and encouragement thought the National Quality Award process. As a Xerox employee, I was pretty “stoked” when we won in 1989. Hope the DOE can be as inspiring.

  7. Students are more aware and quick thinker but maturity lack’s so teacher should be work on that also.

  8. “we intentionally leave that undefined—because different states will have different approaches—and different confidence levels in their assessments.”

    Well, okay, but their approaches determine whether or not they receive Race to the Top $. And are states ever punished for letting standardized testing of students determine too much of teacher evaluation?

Comments are closed.