| Abstract: | The development of a model for more effective usage of group administered standardized tests is developed. The primary assumption is that summary data from these instruments are not used in curricular decisions for groups larger than the classroom unit, and a method is illustrated which uses item groupings from subtests allowing a more effective use of the data for program and curricular decisions. Several cautions are presented with respect to data usage in the recommended fashion, i. e., item data usually have low reliability coefficients, and criterion levels which identify problems in the curriculum are difficult to establish. |
