I have a process for looking at the data each year, a process that informs my instruction. Here's what I do:
- Data Overview: First, I give the data a quick overview to see if any student scores much differently than I expect given daily observation, classwork, formative assessments and other information. If a child scores very differently than I expect, I will look at his/her data closely to find out what happened. I will also refer to notes I keep related to children's demeanor, attitude and events at the test taking time. For example, if I know a child is exceptionally nervous during test taking, sick or encountering a personal issue, I note that as I know it could affect the test outcome.
- State, System, School Comparisons: Next, I look at the data related to State, system and school scores. I notice if my class's scores are much different than those of my peer teachers, teachers in my system and teachers in the State. If the scores are different, I look deeper. I think about how my instructional practices and class composition might differ from theirs. I often make instructional changes after this investigation.
- Question Type Analysis: After that, I look at the specific question types. Overall are there question types that my students do very well on or questions that are challenging? A couple of years ago, we noticed that our students were not doing very well on open response writing in response to text. Our school system analyzed the issue and learned a lot about what the test expected and what makes a powerful open response answer. We also learned a lot about the process, and led children through many meaningful exercises in practicing that skill, a skill that they'll use again and again throughout life--writing a response with evidence from the text to prove or explain a point.
- Content Area Analysis: Then I look at content areas--are there particular content areas that my students do exceptionally well on or areas that challenge them? If I notice that my class in general is challenged by one content strand, I revisit my teaching strategies for that strand, talk to colleagues and research ways I can better teach that content.
- Individual Student Assessments: I also look at my individual students. Which ones perform well, and which students face challenges? If a child doesn't perform well, I think about his/her overall profile and the supports in place. For example, I had one student in the past that did not score well, but she scored better than in the past and many supports had been put into place for her steady progress. On the other hand, I had another student who struggled. He came to our system late, and had very few outside supports. Right away I alerted administrators and teachers about this and advocated for greater support and analysis of the child's learning. Also, if I see a group of similar students struggle in one of more areas, I'll think carefully about those students and their access to teaching. For example, I recognized that some of my less verbal students one year did not make the same progress as my verbal students--I decided to slow down some of my teaching events to make more time for my quiet students' responses and discussion.
My school system also does extensive analyses of the data and makes curriculum and professional development decisions based on the overall profile As I've written before in the post, Standardized Tests?, I'm in favor of streamlined, standardized tests as one of many criteria used to inform and lead our instruction as we develop successful schools.
I'd like to see the tests leveled so that students take the test they're ready for rather than a test based on grade or age, and I'd like to see the tests given on Saturdays with outside proctors rather than their classroom teachers--the people they rely on to help them, guide them and answer their questions when needed. I also hope that the Common Core will include a more manageable number of standards, since we simply run out of time sometimes to include it all.
How do you utilize the scores from standardized tests? Do your analyses differ from mine? I welcome your comments.
