Score! Connecting test results to big gains in your classroom
by Christopher Maricle
When your class's standardized test scores arrive, do you know how to use them to improve teaching and learning? Learn how to interpret test results and how they can help you and your students.
What does it mean for a teacher to be data-driven? Teachers are busy people with lessons to plan, papers to grade, material to organize, trips to coordinate, and more. And that’s after all the teaching! So, when the standardized test scores come, what would really help is a few simple guidelines with specific recommendations on how to use the results in a way that’s both manageable and meaningful.
Before you dive into the test scores, there are three things that you should know:
1. You matter a lot. Research is clear: an effective teacher makes an enormous difference in student achievement. The value of your work is the single greatest factor (of those that we can control in schools) in student learning. What you do is more important than standards, curriculum, textbooks, assessment, homework, parent involvement, or technology. You are the heart of learning for your students, and what you do matters.
2. Data matters. The effective use of data to inform teaching is one of the single greatest research-based attributes of schools that have been successful in improving student performance and closing the achievement gap. Using data works.
3. It is not a difficult matter. I won’t lie and say it’s easy. It does take time, thought and effort. But it’s not hard. You can do it, and you should do it, and you should do it with your colleagues. If you think you don’t have talent at it, don’t worry. Talent, as a wise friend told me, is merely “an interest pursued.” Just start. You’ll get better at it.
Getting Started: Where Scores Come From
The standardized test your students take is very likely a norm-referenced test, designed to compare the performance of students to other students. If so, then the extent to which you can make criterion-referenced interpretations will vary depending on the test. But there are three things you should do before drilling down into the test results.
1. Read the test publisher’s guidance on interpreting tests scores. They’ve gotten better over the years at making these instructions more teacher-friendly and more accessible. To prepare for this article, I performed an online search for one of the test publishers that many Catholic schools use and found the teacher’s section on interpretation of scores in about 30 seconds.
2. Be familiar with the basic structure of the test. For example, in the math test for your grade level, how many sub-sections are there, and how many questions are in each section?
3. Know the content standards for the subjects you teach. This is will be important later, because low scores are the result of only two possibilities (assuming the test is valid, and the student’s performance on the test is an accurate reflection of ability). Either the students are not being taught the content that is measured by the test, or the content is being taught but the instructional strategies aren’t working. You’ll want to know which.
Which Scores Should You Look At?
Your students’ scores are probably provided in several formats. Most commonly these are Raw Scores (RS); Percent Correct (PC); Grade Equivalent (GE); Scaled/Standard Score, and Percentile Score (PR). Some also offer Normal Curve Equivalent (NCE). While each of these has some use, several are fairly limited. For teachers, the one is that is often considered to be the most useful is the GE: the Grade Equivalent score. GEs are great for two reasons.
1) They are very intuitive. The digit to the left of the decimal is the grade, and the digit to the right is the month. For example, if Alicia in the 5th grade gets an overall score of 6.3 on the math test, that means that Alicia scored as well as a 6th grade student in the third month of school would perform on the 5th grade math test.
2) Because the score numbers are tied to grade levels, it makes performance over time relatively easy to measure. A change from a 4th grade score of 4.4 to a 5th grade score of 5.4 would indicate that a student gained approximately one year of learning (5.4 - 4.4=1).
If you have them, Normal Curve Equivalents can be of benefit because they take the percentile scores and divide them up into 99 equal units. That’s helpful because in percentile scores, a 10 point gain from 45 to 55 is not as much growth as from the 5th to the 15th. Percentile points mean more growth at the extremes of the student performance curve. But NCEs account for this, and so five points growth is always five points growth, no matter where it occurs in the curve.
Class or Student Scores: Which Is Better?
Don’t worry too much about class averages; they don’t tell you very much. Here’s why: Pretend that you teach habitats in science. You taught rain forests in the fall, deserts in the winter, and oceans in the spring. Alonzo got an F in rain forests, a C in deserts, and A in ocean systems. So, for the year, Alonzo has a C in science. When his parents see this C, they are likely to believe that he has an average understanding of the science content, but that isn’t the case, is it? Alonzo has a wide range of understanding on a variety of subtopics under the general concept of habitats.
Class average test scores are the same. They provide a very general sense of how the kids in your class are doing as a whole, but the class average score reveals nothing about how each of the students is doing.
What you really want is the range of scores. Organize the student scores from high to low in each subject area in which any students are identified as not performing at grade level. With scores displayed in this manner you will see the range of skills among your students. You will see how many natural groupings you have and the relative number of children who are under-performing relative to the standards.
Go Beyond the Scores
A test given several weeks before you see the results is what we call a lagging indicator. In other words, they provide after-the-fact, can’t-change-the-results data. It is the output side of data. The only way to change the scores next time around is to change the input data—the leading indicators—content, instruction, and benchmark assessments.
Align the Content
For each subtest in which students are not achieving at grade level, you must discern what content is being tested by each item in the test. Compare the content in the test to the content standards for your grade and subject. Are your students expected to master these standards by the time the test is taken? If not, no worries. Here’s an example of why you must know the standards being tested and the standards in your school’s curriculum.
Each year my 5th-grade students performed poorly on fractions. Each year the principal would point this out to me. And each year I would explain that we were testing in the fall and I taught fractions after January. But the 6th graders performed satisfactorily on the 6th grade subtest in fractions. So, I was not worried about fractions, and I did not need to consider changing how I taught fractions because it was working.
Align Your Assessments
There is only one simple question you must answer: Would student performance on your benchmark assessments accurately predict student performance on the standardized tests? If the answer to either is “No,” then there are likely two reasons.
1. The content is not aligned.
2. The instruction did not lead students to the level of mastery of the content.
Develop more accurate assessments that measure the content at the appropriate level of rigor. When these are aligned with the standardized tests, you will get the input feedback you need. You will know which students are not mastering the required content.
Align the Instruction
This is an ego-laden topic: teachers often feel they are teaching the “right way.” The truth is, there is no “right way.” Is it possible to adapt your instruction so that more students achieve grade level mastery of content? How do you change strategies?
Two simple rules:
1. Apply the research on effective instructional strategies for the subject matter.
2. Check with teachers who demonstrated success in teaching the content.
This is really not an individual challenge. These questions are the rightful property of the faculty. In a K-8 school, the faculty is responsible for ensuring there is alignment and continuity in the curriculum. In a high school, the math department has this same responsibility for mathematics. In order to have a cohesive curriculum, the 4th and 6th grade teachers must know and understand what 5th graders need to know and be able to do. The responsibility for cohesiveness and coherence is a collective one.
Having these discussions requires a level of professionalism that is quite demanding. It may ultimately require a teacher to bring a problem to the team by saying, “My students didn’t do well on this section in math on the standardized test. Here is the content I presented last year, the tests I used, and here’s how I taught it. Now, what can I do to improve student achievement in this area? What do you think?”
Sustaining these kinds of conversations is the heart of effective professional development. If you do it consistently, you will grow as professional educator. When you take the output data of standardized test, and compare it with the input data of content, assessments and instruction, you will bring life to the mission language we so often hear: data-driven and continuous improvement. Doing this together as a faculty will make your school stronger and serve your students and their parents in manner of which you will be rightfully proud.
Christopher Maricle served as a Catholic-school teacher, principal, and assistant superintendent from 1986-2006. His book The Jesus Priorities: 8 Essential Habits was published in October 2007. He is a senior consultant for the California School Boards Association. He can be reached at email@example.com.
Copyright 2015, Bayard, Inc. All rights reserved. This article is protected by United States copyright and other intellectual property laws and may not be reproduced, rewritten, distributed, redisseminated, transmitted, displayed, published or broadcast, directly or indirectly, in any medium without the prior written permission of Bayard, Inc.