The end of the year is upon us! Teachers are packing up their classrooms, students are getting their yearbooks signed, and the state test that they took a few weeks ago is probably the last thing on most people’s minds.
Unless you are me.
Most schools have already received their preliminary data, and I love to analyze it right at the end of the year as a self reflection practice. Last year, I talked about how I compare the scores to my grade book to determine whether my grades are accurate. This year, I want to talk about a different predictive tool: The interim test.
Teachers have access to interim tests that they can give throughout the year to measure formative progress. There are two types of interim tests, and I prefer to use the block based tests that assess a particular skill sets. In 8th grade math, there are tests on geometry, functions, expressions and equations, the number system and even a sample performance task. And while I am only required me to give two of them, I give all of them.
Why? Because they are extremely predictive.
In the graph above, I have compared my current students’ averages scale scores on their interims this year with their final preliminary score. The correlation is strong and there is an r-squared of .73, meaning 73% of the variability in their final scores can be explained by their interim scores.
If interims are so predictive, why aren’t they more widely used?
I don’t think it’s well known that interims can be given any time by any teacher without administrator approval. Students can take each interim block twice per year, and they don’t have to be scheduled. The only thing students need is a computer and their code and they can take an interim.
Furthermore, the interim reporting system was significantly improved this year to include item-specific analysis, so teachers can use the test to analyze performance task-by task.
But I think what is equally difficult about using the interims is HOW you use them. Interim blocks are called formative tests because they do not test on all the standards. So in theory, you should give them throughout the year after you teach that subject. However, most curricula don’t teach the standards in the order that the blocks are presented. They follow whatever logical order makes sense. So if you give an interim in November, you may have covered some questions but not others. Often, giving the interim in November is pretty futile.
So here is what I do:
I finish the curriculum by March. I teach all the standards by March. Then, I start to review. I go back week by week covering one set of standards at a time. I then give one interim every Friday. I use that data to pull small groups or differentiate instruction the following week. Every week another interim. Boom Boom Boom Boom and suddenly I have a built in measure of how my final review is going. I don’t give any other tests after march, only interims – so I am not over-testing them, I am just testing them more strategically.
We are teaching in a time where we have reacted strongly against “teaching to the test”. And rightly so. But using these interims to support learning is not the same. These are strong tests that accurately predict mastery. When we use these interims correctly, it means that we are using the data points we have to support student learning. We aren’t teaching to the test, we are testing so we can teach.