by Dr. Barbara Plake, psychometrician
Issue 1, 2008
Dr. Plake is a psychometrician and emeritus distinguished professor at the University of Nebraska-Lincoln. Her research focuses on classroom assessment practices, computerized testing and on methods for determining the passing score on high-stakes tests, such as those for high school. Before becoming an academic, she taught Grade 9 math in Florida.
At December’s EQAO conference, Dr. Barbara Plake, a psychometrician and emeritus distinguished professor at the University of Nebraska-Lincoln, presented crucial criteria and a series of indicators to help teachers prepare and deliver classroom tests that are first-class.
Recent research has shown that good classroom assessment practices can be linked to increased student performance. By helping teachers understand how their students are learning, good classroom tests enable teachers to reflect on and adjust their teaching practices—a prerequisite for improved student learning.
But not every test is a good test. In order for classroom assessments to be a sound source of information on student learning and bring about substantial student performance gains, classroom tests need to possess certain indispensable features.
While the curriculum tends to be the primary focus of instruction, classroom assessments are often regarded as a secondary element in the learning process. However, recent psychometric studies have shown a consistent relationship between good classroom assessments and important student performance gains.
“Research has shown that there are substantial learning gains for students when classroom instruction is strengthened with classroom assessments,” says Dr. Plake. “Most interestingly, these gains are of a greater magnitude than those stemming from the implementation of combined educational reforms, and this should reinforce teachers in the work they do every day. The gains observed are also more notable among students considered at risk— those who have been struggling for years and have been impervious to most other instructional innovations.”
“These research findings are wonderful news,” says Dr. Plake, “but significant student performance gains won’t happen with the use of any classroom assessment.” In fact, research has shown that classroom assessments need to have three key features in order to be of good quality and bring about substantial student performance gains:
In this phase, the focus is on the development of the test, as well as on its scoring and the interpretation of the results.
Out of the three key features of a good classroom assessment proposed by Dr. Plake, providing accurate information is certainly the one that takes the most time as it involves test design, scoring and grading. To help achieve that, Dr. Plake defined six student-metric indicators. These indicators and the questions that they raise will help teachers deliver a classroom assessment that will provide accurate information about what students know and what they don’t know.
For example, for a Grade 8 reading test it is good to have some passages that are a little higher and others a little lower than grade level. But the test cannot have Grade 3 or Grade 10 material as that would be inappropriate. It is important to calibrate the reading level of the test.
It is important that the reading level of the test not interfere with students’ ability to demonstrate that they know the skill or concept assessed on the test. Reading level is considered one of the extraneous factors that can impact student performance and should, for that reason, be minimized. Cultural differences, student background and prior knowledge are among the other extraneous or confounding factors that can affect student performance and lead to inaccurate interpretations about student achievement in other content areas, such as science or math.
It is important to reflect on this question as a score of 90% on a test could be very easy to achieve and not show a high level of performance. On the contrary, a 50% score may be very hard to achieve and show a high level of performance. Here are a few suggestions to circumvent this “illusion of equivalence.”
Classroom assessments are used to fulfill a wide variety of objectives, from communicating student achievement and evaluating instructional effectiveness to making instructional adjustments and reporting achievement results for accountability purposes. These objectives are not suitable unless classroom assessments display the requisite level of technical quality that ensure they provide accurate information about student learning. “Teachers are the linchpin in making this work because they are the ones who know the students, make the decisions and deliver the instruction,” concludes Dr. Plake. The lessons to be learned from good classroom assessment practices are tremendous and the rewards are invaluable.