Teacher Involvement to Evaluate Achievement

Much is written about using standardized and state mandated tests to measure student growth in science. Little, however, is written about using teacher developed tests to measure student science achievement in the classroom. Too frequently, standardized tests and state mandated tests omit learner progress in the every day learnings in science achievement in the classroom. Teacher developed tests can be more valid and reliable then either standardized or state mandated tests. Standardized tests are global and do not have accompanying objectives for the teacher to use in teaching science. State mandated tests may have accompanying objectives which teachers may use to gauge their teaching. The teacher may then teach in a manner whereby the state mandated test might have rather high validity since what is taught may' be tested when using the state mandated test. Selected states omit science content and skills in the mandated tests. The feeling then might well be that science is not as important as reading and mathematics, for example. There is probably no better way to write valid test items than the classroom teacher doing the writing who knows and under stands what has been taught in the classroom. The science teacher also has understandings of what the developmental level of the student is when writing appropriate test items. Thus, test items need to be written on the reading and understanding level of the students who are to take the test. Too many teachers, however, are not acquainted with standards to use in test writing. Standards for Test Writing The National Research Council (NRC) has just released publication of Classroom Assessment and the National Education Standards (National Science Teachers Association, 2001). Six goals are listed in the introduction of this publications and these are the following: * Articulate a research based rationale for helping teachers improve classroom assessment. * Clarify the concept of effective classroom assessment. * Provide illustrations and guides to the development and selection of assessment processes and tools. * Assist teacher educators and staff developers who will include assessment in their work with prospective and practicing teachers. * Address issues that school and district decision makers face in their efforts to improve classroom assessment. What students learn each day in ongoing science lessons and units of study is important. A single percentile given to show student results on a standardized or state mandated test for an entire school year, or other selected interval of time, does not provide a comprehensive picture of the learner's achievement in science (Ediger, 1990, 241-246). To use either of these tests to reveal learner progress in science has the following weaknesses: 1. multiple choice test items are generally used for student responses to indicate what has been learned. These are paper/pencil test items to measure facts, concepts, and generalizations acquired by students. Major emphasis may be placed here upon measuring factual information achieved by students since this is the easiest kind of content to measure in a multiple choice test. 2. hands on approaches in science teaching and learning might well be greatly minimized or ignored since their strengths and weaknesses cannot be machine scored. A science teacher who has stressed a hands on approach in student learning in the classroom may find the multiple choice test items tend to measure verbal intelligences such as reading largely or only. 3. no exceptions may be made for individual differences among students. Thus all students have the same amount of allotted time to complete the standardized or state mandated test, all experience the same directions for test taking, all responses are assessed using the same answer key, and all respond to the same kind of test items. Selected students need more test taking time than do others as well as to have directions explained more thoroughly for taking the test. …