BC Education - Mathematics 8 and 9 Appendix C

Appendix C: Assessment Practices

Teachers should base their assessment and evaluation of student performance on a wide variety of methods and tools, including observation, student self-assessments, daily practice assignments, quizzes, work samples, pencil-and-paper tests, holistic rating scales, projects, oral and written reports, performance reviews, and portfolio assessments. Using a variety of assessment methods can help teachers to compile comprehensive profiles of student learning. The Assessment Handbook Series -- Performance Assessment, Portfolio Assessment, Student Self-Assessment, Student-Centred Conferences, and Numeracy Performance Standards -- provide useful, detailed information about a range of appropriate assessment practices. The rest of this appendix provides guidance in creating classroom tests.

Constructing Classroom Tests

There are two types of evaluations that take place in a classroom, each with its own distinct purpose.

Formative evaluations are ongoing assessments used to guide learning rather than draw final conclusions.
Summative evaluations refer to evaluative procedures (e.g., tests, reports, projects) usually conducted at the end of major units, to assess performance against predetermined criteria. These normally represent a substantial portion of final grades. The classroom tests described in this section fall under the category of summative evaluations.

All tests should be developed using the principles of criterion-referenced evaluation. In criterion-referenced evaluation, student achievement is interpreted in relation to previously defined levels of achievement, rather than relative to the achievement of other students. Questions on a criterion-referenced test should be representative of a clearly defined domain of prescribed learning outcomes. In this way, scores more accurately represent the student's present status with respect to those outcomes.

A test should measure what it intends to measure. For example, if a test requires a reading level far above the abilities of many of the students taking the test, test results would measure differences in reading levels rather than differences in subject knowledge.

Steps in Classroom Test Construction

The following suggests the key points to consider when developing classroom tests.

Steps in Classroom Test Construction	Points for Consideration
Plan the Test	Begin the planning process well in advance. Identify the prescribed learning outcomes to be tested. Learning outcomes provide a framework for the development of criteria. Create a table of specifications that covers the learning outcomes and cognitive levels (knowledge, understanding, higher mental processes). Balance the table of specifications to reflect the topics and cognitive levels in the curriculum.
Write Test Items	Word questions clearly. (e.g., Use "determine a value for x" rather than "find x.") Define the answer format so students understand the form their answers must take. Don't repeat questions on one outcome. When possible, design questions that connect various topics within the curriculum area and between various curricula. Create questions that require various forms of answers (e.g., explaining, comparing, illustrating, graphing, calculating, solving, justifying). Categorize every question according to the criteria. Don't phrase distractors in multiple-choice questions so that they point to the answer. Review questions for appropriate vocabulary and targetted reading levels. Ask a colleague to answer the questions to identify possible marking problems and time requirements, and to provide feedback.
Format the Test	Use easier questions at the beginning of the test to establish students' confidence. Group similar items. Organize items on the page so they are easy to read and provide adequate space for responses. Develop test instructions that are clear and unambiguous.
Develop a Scoring Key	Mark for processes as well as correct answers. Allow for possible alternative solutions or different forms of answers (e.g., format, notation, detail). Consider a range of scoring methods (e.g., holistic, analytic).
Prepare Students	Establish test criteria with students. Help students brainstorm topics that are likely to be covered. Discuss how to approach the test (budget time and the weight given test results in the final grade). Give students sufficient time to prepare for the test. Review terminology used on the test (e.g., evaluate, simplify).
Administer the Test	Allow time for all, or nearly all, students to finish. Ensure a test location free of distractions. Ensure all necessary supplies are available.
Score the Tests	Test the effectiveness of the key on a few sample tests. Adjust the key as necessary. Note exemplar papers. Mark tests at one sitting (or one question at one sitting) to ensure consistency. Return marked tests promptly, and review them with the students to help them improve their understanding of the concepts involved.
Table of Specifications	A unit test in mathematics must measure the skills or concepts taught in the unit. A table of specifications can help the teacher plan the amount of emphasis to give each skill or concept. A table of specifications is a chart showing the content categories and cognitive levels to be tested. The percentage weighting of test items within any row or column is determined by the amount of time devoted to the content categories and their degree of difficulty.

Table of Specifications Unit #________ Variables and Equations Principles of Mathematics 9
Content	Knowledge recall conventions classification notation	Understanding application of theories, ideas, principles, or methods to a new situation	Higher Mental Processes analysis synthesis evaluation	Percentage of Total
Problem Solving		3 questions	4 questions	28%
Algebraic Skills	5 questions	6 questions	2 questions	52%
Mathematical Reasoning			5 questions	20%
% of Total	20%	36%	44%	100%

Revised: September 1, 2001

BC Ministry of Education Home Page