For Trainers Only
Module 3 Guidance:
In order to help participants understand the different ways that growth measures can be structured, the facilitators need to explain how each of the growth target examples build on each other by addressing the variety of growth rates within student populations and/or improving the precision of the growth targets. The growth targets build on each other, becoming more and more detailed in their differentiation and potential for precision.
When starting this discussion, clarify that the term “precise” doesn’t always mean “better.” If a teacher has solid information about what student growth to expect, then a more precise target could be better. But “precise” doesn’t mean better if the information used isn’t grounded in knowledge of the appropriate student growth to expect. Precision and accuracy are required for quality growth targets.
Below are a series of points to make when discussing each target example.
Target 1:
· This target is structured so that ALL students are expected to show growth and that growth is based upon where students are currently performing (i.e. Zone 1 would grow to Zone 2, Zone 4 would be expected to grow to Zone 5). This is a good thing.
· This target assumes that one tier or performance level of growth is challenging and appropriate for all of the students in the population.
· This target also suggests that moving from one tier to the next at the bottom of the performance spectrum is comparable to moving from one tier to the next at the top of the performance spectrum.
· Further, this target assumes all students have at least one zone to grow.
· “Zone” is referenced here specifically as a part of the example assessment. Please avoid using the term zone in other examples when referencing SLO “tiers” as it may be confusing to participants.
These aspects of target 1 make it a simple growth target that may be applicable in only rare instances. For example, this goal makes sense when students will all grow at the same rate, where growth to only one level above is both challenging and possible, and where growth on the assessment is even and consistent (i.e. moving from one tier to the next means equal amounts of learning). Since these conditions are rare, we offer other examples that are more differentiated and precise.
Target 2:
· This target is one example of how you can set differentiated targets for the students covered in the SLO based on a simple formula. It’s valuable because scores are differentiated to individual students based on their pre-test scores.
· This example assumes that students who begin with more knowledge and skill will demonstrate less growth in their score than their peers. For example, a student who scores a 80 on the pre-test only needs to score 10 more points to achieve a 90 on the post-test. A student who scores a 50 on the pre-test needs to score 25 more points to achieve a 75 on the post-test. Some assessments will be built such that growth is more difficult to demonstrate at the higher levels of performance, suggesting that these examples are sound, but this is not always the case.
· As teachers and grade levels collect long-term, trend data on their students, they can analyze data and set more precise targets that have a better chance of being accurate for students.
Austin has been working on developing district assessments that are similarly structured and comparable across subjects and grades so using a formula may be appropriate. In Ohio, keep in mind that teachers may be using a variety of assessments that are structured differently so that using a formula may not be appropriate.
· Please note: ODE will not develop a growth formula for a district. However, LEAs can decide to set a growth calculation, but the following assumptions apply to make these growth targets tenable.
Target 3:
· This target is an example of setting growth targets in tiers of performance levels based on trend data. It’s a good example – but it has some limitations.
· Invite participants to consider the amount of growth a student who scores a 60 will need to make.
· Then invite participants to consider the amount of growth a student who scores a 61 will need to make.
· In this example, for students who score close to a cut-point, there is wide variation in the degree of growth required depending on which the side of the cut-point they fall.
Target 4:
· Target 4 addresses some of the issues addressed in the review of Target 3. Here, growth expectations differ, but do not differ as dramatically, around the cut points.So, for example, in target 4, a student who scored a 60 is expected to grow 15 points and a student who scored a 61 is also expected to grow 15 points.
· Target 4 sets a minimum score as well as a gain score required to meet the growth goal.
Target 5:
· This example is unique in that it is a descriptive growth target. The goal here is to show an example demonstrating that not all targets have to be numerical.
· Point out that the descriptors were developed through content experts (Linguafolio). If you were to create your own rubric and use that as part of your growth targets, content experts would have to vet the rubric and make sure that the expected jump across rubric levels is appropriate.
· Note that the mathematical formula used by Austin Independent School District would not work in this case.
Know your students, know your assessments:
When setting growth targets, the teacher must consider how the assessment is structured and the nature of knowledge and skill development required by students when they take the assessment. Based upon the assessment structure, what is an ambitious yet attainable expectation? Some assessments may be structured so that a uniform growth expectation (i.e. all students will increase their baseline score by 20 points) is possible. Other assessments may be structured so that lower-performing students can reasonably be expected to show greater growth than their higher-performing peers. Information about your assessments and information about your students will help you select an appropriate format for your growth targets in your SLO.