Growing With NWEA

We have decided to engage the services of the non-profit organization, Northwest Evaluation Association, to provide assessments that measure the growth in achievement among individual learners. The three tests that will be administered (fall, winter, and spring) will afford us with specific data that we will be able to convert into informed decisions that drive subsequent instruction. These tests are simply referred to as growth models in that they measure and monitor progress from point to point.

Let me explain the primary difference between these tests and the state mandated exams that learners across New York State encounter each spring - as it relates to the teacher evaluation component of the Annual Professional Performance Review. The state tests produce an outcome in the form of a static reference point. For example, each child receives a score from 1-4 (3 and 4 are required for mastery) on the basis of a scaled score rendered by the number of correct answers. For purposes of evaluating the teacher, the teacher is thus measured by the percentage of learners reaching a 3 or 4 on the test.

However, this format does not measure the full impact of the teacher. Let's take a very bright learner who arrives in 4th grade with the capability of taking the test on the first day of school and receiving a 3 or 4 on the exam. They are already at the desired level but sit through the class acquiring seat time until the tests arrive and they pass it. According to the child's 3 or 4 on the test the teacher is perceived as effective - and if enough learners receive a 3 or 4 then the teacher may be considered highly effective. Yet, the teacher really didn't necessarily influence the performance level of that particular very bright child.

On the other hand, let's look at a another 4th grade teacher who welcomes someone into the class who is an underperforming learner - someone who receive a 1 on that same test if it was given on the first day of school. The teacher consistently exercises strategic instructional practices that leverage success and at the end of the year when the state tests are administered, the child receives a 3 or 4 on the test.

Are these two teachers, who produced the same performance outcomes on the same assessment instrument, equal in the impact they had on the learners cited in our example? This example raises at least one criticism of the manner in which the state tests are employed in making important decision on evaluating teachers.

The growth model of tests determines the starting point for each individual upon taking their first test (in the fall) and uses that as a baseline for instruction. Specific data is generated for each learner and demonstrates where they are on a continuum of skills, with instructional prescriptions emerging from the results in the form of what skills must be addressed next to mitigate deficiencies. This process is repeated at the next testing interval (in the winter), providing the teacher with meaningful and relevant information on progress. Finally, the third test, given in the spring, will yield data that can then be measured against the individual's original starting point to determine the extent of growth evidenced by the learner and orchestrated by the teacher. That comparison more accurately reveals the degree and impact of the teacher's intervention techniques than the use of a single data point arrived at from an end of year test that lacks any baseline but the score from the year before (which is difficult when the state changes cut scores that differentiate between 3's and 4's).

We feel that both learner and teacher will likely benefit more from the use of the growth model test pattern because it uses three different data collection points and allows the teacher to utilize the information produced in this year long progress monitoring cycle to inform instructional decisions.