A NEW LOOK AT ASSESSMENTS: THE STUDENT PERFORMANCE GRID

By Amy Colton and Georgea Langer (a Corwin Press blog)

In our book The Collaborative Analysis of Student Learning, my co-author and I explore many powerful protocols for analyzing student assessments. One of those is the Student Performance Grid, a powerful tool for looking deeply into a class set of assessments. The grid requires a detailed description of each student’s performance and thus presents an accurate picture of what patterns actually exist.

Most teachers score their student’s work and then record the grades. A major pitfall to this “score and grade” approach is the assumption that a single score tells the whole story. But a score of 2 for one student may not mean the same as a 2 for another student. Although some students have the same score, the reasons for those scores are often different. For example, Joe might have a low score because of lack of organization in his writing, whereas Sue earned the same score because of poor mechanics. Such differences often call for different teaching approaches.

The Student Performance Grid protocol (described below) is one of many tools used by teachers to analyze student work while engaged in the Collaborative Analysis of Student Learning (CASL). The Grid provides teachers the opportunity to look beyond the score to find evidence in the work of the student’s specific misconceptions, strengths, and ways of learning. The resulting information is used to establish learning goals for themselves and their students.

Teachers who use the grid experience its power for analyzing each student’s performance in detail and as a result, they often choose to use this method for other important assessments.

The grid is most effective when teachers use an assessment that measures multiple aspects of performance using an “analytic rubric” with more than one category of performance. For example, in writing the rubric might require different scores for the categories of Organization, Voice, Supporting Detail, and Conventions.

Student Performance Grid Protocol

• Sort the class set of student work into the number of scores on the rubric (e.g., three piles for a 2-level rubric). Take notes (to share later) about which parts of the rubric were hard to use or confusing.
• Preparing the Grid.  Take a sheet of paper or start a table on a computer. Draw 4 columns and label them “name”, “score”, “strengths”, and “areas for improvement”.
• Start with the pile with the lowest scores. Enter each student’s name on the left side of the grid, record the score, and then write specific details you see in the paper–both the Strengths and Areas for Improvement: Use more detail than the rubric provides. Sometimes, crucial clues to student learning may appear in the work, but may not appear in the rubric.
• Data: Enter every low-scoring paper onto the grid. Then go on to the next pile (middle scores) and to the same. Continue until all papers are entered, arranged from lowest to highest.
• On the grid’s “areas of improvement” column, color-code common patterns.
• Common Patterns. On the Performance Grid, place a star (*) next to the two or three most common color-coded areas.
• In a group with other teachers, dialogue about the different patterns found in one another’s grids to help decide which patterns might be the most valuable to study, using a question such as, Based on the data, which learning outcomes are most in need of improvement?