Some Techniques for Determining the Relative Magnitude of Speed and Power Components of a Test
Abstract
Some Techniques for Determining the Relative Magnitude of Speed and Power Components of a Test SAGE Publications, Inc.1953DOI: 10.1177/001316445301300211 Gerald C. Helmstadter University of Minnesota Dale H. Ortmeyer Columbia University WHEN a function which a test attempts to measure involves factors of both speed and power, the validity of the test will depend to some extent on the relative importance which speed and power elements have in determining the scores made. Likewise, the split-half methods of estimating test reliability (and especially the odd-even) are justifiable only when the speed component of a test is quite small. Further, although any test may rationally be considered as largely speed or largely power, the relative importance of these two components is not independent of the group being measured, and a test which samples depth of ability for one group may be measuring only a speed component for a second. Thus, in the evaluation of a test, it is often desirable to know precisely the relative parts which the speed and power elements play in determining the scores for some particular group. It is the purpose of the present article to present several methods for doing this. The Techniques Before a method