By Mary Anne Mather, Managing Editor
TERC’s Using Data For Meaningful Change Blog
TERC Using Data Senior Facilitator

Many districts are heading into spring state-level testing. It’s irrefutable that the opinions surrounding the pros and cons ofthree teachers collaboratively analyzing student work samples such assessments make for heated discussions in many circles. Not the least among the disputes is the time spent on what some call “teaching to the test.” The high stakes value placed on these tests can make even the best of us do things we don’t really embrace as best practice.

At TERC, we try to look at it from a different angle. What if our day-to-day work as professional-level instructors set the stage for students to perform better on the standardized tests because we intricately understood the ins and outs of what students do and do not know? Armed with that knowledge, we can plan classroom instruction that closes the gap between misconception and success. It’s most likely going to influence test scores, while addressing essential grade-level learning goals. That’s where looking at student work samples comes in!

Knowing the Learning Challenge
Throughout the school year, effective teachers are constantly consulting multiple data sources to pinpoint areas where individual or groups of students struggle. As seasoned educators, we note trends in benchmark assessments, end-of-unit tests, and day-to-day classroom performance using formative assessment strategies. By this time in the year, we are pretty well aware of the standards, strands, and broad concepts that challenge students the most at a specific grade level. And these are most likely among the standards that will be evident in the state test items.

Preparing For the Task
Now is the time to come together as a team and draw on one another’s expertise to unpack that standard, strand, or broad concept and select an open-ended task related to the challenge area that will really test a student’s content-understanding moxy. The best tasks are open-ended and offer a level of cognitive demand that requires lots of thinking. An open-ended task gives students freedom about how they might approach the task and allows for multiple avenues to success. When students undertake the task, they are required not only to give their answers, but to show their work and explain their thinking. That’s where the rubber meets the road.

Most likely, the selected task reflects content that was already taught, but not mastered by all students. You can help students prepare for the task by offering similar open-ended task samples and facilitating individual and group think-alouds about the solutions. You can also show your students exemplars of mastery or examples of incomplete answers and ask them for suggestions about how to improve the answers. Then, follow on with asking all students at a grade level to individually solve/complete the selected assessment.

Scoring the Task
In TERC Using Data workshops, we ask each teacher in a grade level or Data Team, before scoring the student work, to complete the open-ended assessment individually—just as if he or she were a student. The real value in this comes as teachers share their answers and solutions with one another, and they “deconstruct” the task. They engage in a collaborative dialogue around what knowledge, skills, concepts, and big ideas are necessary in order to master the task.

This task deconstruction serves two purposes:

• It is a type of professional development that reveals different levels of understanding and varied ways of thinking among peers.

• For all intents and purposes, it results in a collaboratively-developed scoring rubric that will be used to review student work. Teachers come to a common understanding about students must know and be able to do in order to successfully complete the task.

three samples of student work on a math taskThen we ask the team to identify 6-8 pieces student work selected from across the grade level. Student and teacher names are removed from the samples, and the team works together to score against the rubric created as a part of the task deconstruction conversation. Samples should encompass multiple proficiency levels to really get at trends about student misunderstandings or lack of knowledge. During the scoring discussion, look for evidence of student thinking. Are there knowledge gaps, computation challenges (math), weak skill areas, vocabulary misunderstandings, incomplete answers? After the sample work is collaboratively scored, teachers can review each piece of student work for their individual classrooms, using the samples as exemplars.

Action Planning
The most effective next steps include:

• Determine if specific misconceptions are widespread or involve groups of students or only individuals. This lays the groundwork for instruction choices: large group re-teaching, setting up strategies for small group review, or conference sessions with a few individual students.

• Work as a team to share teaching strategies. If many students in one class did well, what did hat teacher do? Discuss new approaches to teaching concepts already covered. Consider teaming—with different teachers working with small groups based on need.

Ultimately, it’s not about teaching to the test. It’s about teaching to student needs. In the end, we believe students will know more, be more proficient…and yes…they even might perform better on “the test.”