When reviewing data with teachers I am sometimes asked, “why not give all students in my class the same test items?” I understand where this question comes from. When you are looking at your students’ test results it makes sense to want to see the questions that most students answered incorrectly so that you can then focus instruction on those areas. This approach does not benefit some students and gives the impression that most students have the same need which often is not the case.
When we give all students in a class a test with the same test questions we are using a “static” or “fixed-form” test. Test designers typically select test questions for a static test at the middle of the grade level in terms of difficulty. The test designer is making a compromise by selecting the most test questions that will be in the learning zone of the most students. If the test has 25 questions maybe 20 of those questions will be designed for the middle of the grade level with a few easier questions and a few harder questions.
For your more advanced students they will likely answer correctly nearly all of the questions on a fixed-form test. You might have even anticipated this before you administered the test and concluded that the test really does not give you any new information for these high achieving students. And the test does not answer the most important question, “what do I do next to help this student progress?” Since the student answered all or nearly all the questions correctly it is not clear where their learning zone might be.
In educational testing this scenario is called the “ceiling effect.” It means that a student hits the ceiling of the test in terms of available test questions that are sufficiently challenging. Not only does a test with the ceiling effect not provide information on where a student may need to go next, it also gives the impression that the student is “making progress” and is not in need of any adjustments in her educational plan.
My concern with the celling effect is that the assessment is not identifying the needs of the above grade level student and as a result those students may not make as much progress as they otherwise could if they were sufficiently challenged or instructed closer to their learning zone. For example, students who test in the 90th percentile in grade three only to test at the 70th percentile several years later are making less progress than students who test each year at the 70th percentile. When we use only static assessments it may not be possible to see the students who are not making expected progress. You can read more about the negative consequences for above grade level students with ceiling effects in this article from Northwestern University.
My concern with the celing effect is that the assessment is not identifying the needs of the above grade level student and as a result those students may not make as much progress as they otherwise could if they were sufficiently challenged or instructed closer to their learning zone.
For students who are below grade level there are also negative consequences to using only static assessments. Similar to the high-flying student, the below grade level students are not finding enough test questions in their learning zone. The below grade level student is answering most and sometimes all of the test questions incorrectly. Not only is this a demoralizing experience for the student but it also does not provide much useful information to you as the teacher. You may have anticipated before you even gave the test that several of your students where not going to answer many if any questions correctly.
In educational testing this is known as the “floor effect” and it presents a specific risk quite different than with the celling effect. When the floor effect is present it may appear that the student needs additional instruction or practice with the specific material on the test. However, the reality may be that the student is one or more grade levels below this material and in fact needs intervention at a foundational level well below the level of the test questions. The double cost for the below grade level student is that not only was the test taking experience demoralizing and frustrating, but the intervention to follow can also be demoralizing and frustrating because it is simply at too high a level. You can read more about the floor effect in this article from leading Response to Intervention researchers.
Computer adaptive tests minimize the likelihood of ceiling and floor effects for your students. For the below grade level student the test gets easier as the student answers questions incorrectly and continues getting easier until the students' learning zone has been identified. At this point the computer adaptive test provides a good mix of test questions in the students' learning zone. For the above grade level student the computer adaptive test becomes progressively more challenging with successive correct answers. Once the test determines the student’s learning zone the test stops increasing difficulty and stays within the student’s zone.
There are a number of important benefits for your students from the computer adaptive approach to student assessment:
- Computer adaptive tests use fewer questions than fixed form tests which means less time testing and more time learning.
- More of the test questions will be in each student’s learning zone resulting in more comfortable test experiences.
- You can get a better idea of how high some students can go and how low other students need to go to be in their learning zone.
- You will often see better motivation and engagement, as its more enjoyable to work within one’s learning zone.
- Students will not see the exact same questions as their neighbor reducing the temptation to “borrow” someone else’s work :-).
- Computer adaptive tests are commonly more accurate than fixed-form tests, providing you with more accurate scores.
- Computer adaptive tests do a better job of capturing progress over time, especially for students below and above grade level.
Track My Progress is a computer adaptive test for students in grades K through 8 in reading and math. You can sign up for a free trial and watch how the test adapts to each of your students’ needs and provides you with important information to guide their learning.