Planning Classroom Tests

Determining the Purpose of the Test

In using classroom tests, teachers may have a broad purpose such as ‘grading’ at the end of a semester, but in some instances they may rather focus on a specific purpose such as ‘whether students demonstrate understanding of mole concept in a non-routine problem setting’.

No matter how broad or narrow the decisions to be made, teacher-made tests are used for various general purposes, as indicated below:

• Understanding entry characterisitics of students,
• Monitoring learning progress of the students throughout the instructional process,
• Understanding the reasons for persistent learning difficulties among students,
• Grading general achievement level of students at the end of instruction,
• Evaluating the effectiveness of instructional processes and materials,
• Providing feedback for students to monitor and assess their learning progress,
• Enhancing student understanding of their interests and progress,
• Identifying students' strengths and weaknesses in terms of different aspects of subject matters they cover,
• Identifying students' strengths and weaknesses in terms of cognitive processes to be developed,
• Providing information for the school administration to convey and implement curricular, extracurricular and counselling activities,
• Selecting students for some remedial or advanced level courses,
• Conducting research studies in terms of students' learning and progress.

In a broader sense, even though assessment results are used for different purposes, the core of the classroom tests focuses on student needs and expectations, and the results are generally used for selection, placement, instructional, pedagogical, and administrative decisions. Focused on instructional processes, classroom assessment is generally organized around placement, formative, diagnostic, and summative decisions (Bloom et al., 1971; Gronlund, 1993; Nitko, 1989).

In the very first step of planning classroom tests, standards to which students' performance will be compared and evaluated should also be determined. Teachers need to specify a minimum competency level either by referencing performance of a group, or setting a minimum standard task that students should demonstrate. The nature of the minimum competency level to be used could be norm-referenced or criterion-referenced, depending on the purpose of testing. For instance, grading may require norm-referenced interpretations; on the other hand, preparing students for a new learning task may require a minimum level of understanding of a set of concepts which could be done within the framework of criterion-referenced testing.

Preparing Test Specifications

In this step three questions should be addressed in order to plan a classroom test: (1) What knowledge such as facts, concepts, principles, generalizations, algorithms, etc., are necessary for students? (2) What content domain is necessary for students? and (3) What cognitive skills and processes are planned to be developed to deal with acquired knowledge and content domain? The answers to these questions are found in the table of specifications (content versus process matrix) where, in a two dimensional chart, it is possible to list all the content of the course, and the knowledge and cognitive processes to be developed in students. In the 1970s Bloom's taxonomy of educational objectives was considered as one of the pioneers in categorizing learning outcomes of students in cognitive domains (Bloom et al., 1971). In its broader sense, Bloom's taxonomy of educational [Page 728]objectives covers knowledge, comprehension, application, analysis, synthesis and evaluation levels. Other taxonomies proposed in the following years have more or less the same idea of the hierarchical structure of human thought. For instance, Williams and Haladyna proposed a taxonomy consisting of recalling, summarizing, predicting, evaluating, and applying steps in developing a thinking schema in line with facts, concepts, and principles (Roid & Haladyna, 1982). However, with the impact of cognitive psychology on teaching and learning, more emphasis was given to complex cognitive skills, and assessment became more process and product oriented rather than tracing individual performances on paper-and-pencil tests (Calfee, 1995). The taxonomy proposed by Royer, Cisero, and Carlo (1993) seems promising and pioneering to deal with higher order thinking skills of students within the framework of cognitive psychology, introducing measures such as: (a) knowledge acquisition, organization and structure; (b) depth of problem representation; (c) mental models; (d) metacognitive skills; (e) automaticity of performance; and (f) efficiency of procedures.

...