Developing Quality Multiple-Choice Tests for Physical Education
Ayers, Suzan F., JOPERD--The Journal of Physical Education, Recreation & Dance
A quality physical education program is one that emphasizes all three learning domains: the psychomotor, the affective, and the cognitive (National-Association for Sport and Physical Education [NASPE], 1995). The psychomotor domain has long been the principal focus of many physical education programs, while the affective domain has been central in more recent work dealing with personal and social responsibility (e.g., Hellison, 1995). It is the cognitive domain, however, that may present the greatest challenge to physical educators. Knowledge in the cognitive domain is a critical part of student performance; therefore, teachers will need to develop the skills to assess that knowledge.
Performance measures (e.g., teacher observation, skill rubrics) have long been a standard component of physical education assessment. While they provide a measurement of student psychomotor skills, performance assessments do not always reflect the cognitive knowledge that students possess. Written tests can fill this void in student assessment. A variety of written test formats can be used to assess learning in the cognitive domain, including essay, short-answer, fill-in-the-blank, true/false, and multiple-choice. Multiple-choice tests in particular can provide a link between performance and cognition that is currently missing from many assessments of student learning in physical education. Although the initial creation of such tests can be time-consuming, multiple-choice templates, once developed, can be easily modified for a variety of instructional topics.
The guidelines presented in this article are designed to help teachers develop quality multiple-choice assessments. Although the entire process of test development is discussed here, many physical educators may not have enough time for full implementation. Thus, teachers are encouraged to apply as many of the guidelines as possible in order to improve the quality of their cognitive assessments.
Some teachers may ask, "Why should I put forth the time and effort required to design a multiple-choice test?" The answer is threefold. First, the benefit of multiple-choice tests is that, once they are developed, they are easy to modify for different units. A little work in the beginning can have big payoffs in the end, including: ease of grading, ability to use items across a variety of content areas (e.g., tennis, badminton, volleyball) with few changes, and feedback on student learning relative to the cognitive level at which instruction was provided.
Second, assessing the cognitive domain can improve the instruction provided in a physical education program. By determining what knowledge students are acquiring, teachers can provide additional instruction when necessary. Assessing cognitive learning also creates a more balanced physical education experience for students.
Third, sharing test-development responsibilities with other physical educators in your school or district can reduce the work required of each individual. If each teacher develops a multiple-choice test for one or two units and shares these tests with other teachers, the school or district can create a bank of test items for the entire curriculum. This type of collaboration can enhance each teacher's test-development skills while creating a valuable instructional tool for the school or district.
It is critical that cognitive tests match the instruction that teachers provide both in amount and emphasis. These issues will be addressed in detail later in this article. The purpose of this article is to offer insights into the process of developing quality multiple-choice tests and to identify the practical uses of test-writing skills.
Interpretive Multiple-Choice Tests
It has long been thought that multiple-choice tests do not provide information about students' higher level of understanding or their ability to apply information in different settings. That belief has been widely refuted in the measurement literature, however (e. …