My school recently instituted interim assessments, by grade teams. I am on the ninth grade team and just finished marking the science section for the entire school. There is also math, english and social studies on the exam. How can the data be analyzed in a meaningful way? I have some thoughts on this, which I am keeping to myself because I would like a wide-range of opinions on this question.
8 Answers | Add Yours
When you look at the data, don't forget to disaggregate it and analyze it by question. By tracking not only which questions were missed but what incorrect answers were chosen, you will learn a great deal about the validity of the test and the instruction that preceded it. Since you are working in teams, hopefully you can trust each other enough to compare different teachers' scores on different sections. It is possible that one person has a really good strategy to share, and you can all benefit from it.
Some of these suggestions seem as if they would only truly be valid if there was a common curriculum in place. Are all the grade level science teachers teaching the same concepts on the same schedule? This would ensure that material being tested was in fact being taught. After this was established, then you could look at class rosters and teachers with regard to the implications of the results.
It may be an uncomfortable thought, but it might be interesting to look at student scores by teacher. If one teacher's scores are higher it may be an indication of the luck of the roster, but it may also indicate that teaching strategies were employed in a way that helped student's achieve the goals of the class. Teacher's have to be willing to learn and share best practices with one another.
If it is an interim assessment I am assuming there was some sort of assessment prior to this one. If so then we could compare those two scores. Also if we were to add a third test at the end of the semester you then have three points of data to compare. I think it is also important to analyze the test itself to see if certain concepts are understood better than others.
I think #4 makes a really important point. If you look at information like this in isolation, it is not going to be as beneficial as if you can look and see how students have improved from their original starting point. Also, if you don't take this factor into consideration, it is potentially damaging to a student to be classified or graded poorly when they might have actually achieved a significant advance from where they started from.
Would it be possible to track value-added performance - ie which students had made the most improvement from a baseline of earlier assessment? This is always useful data for me.
I don't know what's available, but if I were you I would want to get longitudinal data. I would want to be able to track individual students as they go through the school to see how they are progressing. Is that possible with the data that your school is collecting?
When we look at schoolwide testing data, the information that catches my eye as the most useful involves gender, income and ethnicity. That is, we can see if there are trends among specific populations, or if we are more or less successful with boys, girls, Latinos, etc. It can help us to refocus our efforts and adjust our approaches so we can best address the disparities.
We’ve answered 318,960 questions. We can answer yours, too.Ask a question