Comparative Difficulty of Three Forms of Computation Tests

T. G. Foran, Sister M. Albert Lenaway · The Journal of Educational Research · 1938

The first standardized tests in arithmetic computation were com posed of several sections in each of which the examples all involved the same process. Such tests as those prepared by Courtis, Monroe, and Woody employed this arrangement. Woody and McCall introduced the mixed-fundamental form of test in which the examples are arranged in order of difficulty and independently of the process demanded by the calculation of the answer. The mixed-fundamental form of test has been widely adopted and, with slight modifications, it is used in most current tests in computation. Several criticisms of the mixed-fundamental arrangement have been supported by observations regarding the frequency and the types of errors that pupils made. The advantage generally credited to such a form of test is derived from the similarity between the ordinary usage of arithmetic computation and the tasks composing the test. The older form of test was somewhat artificial. It was modeled after the usual drill exercises in which a single process was repeated through many examples. On the other hand, it has been stated that the constant shifting from process to process is not representative of the ordinary use of computation. Such shifting as may be used in problem solving is in response to a con ceptual situation rather than to a symbol or word that precedes the ex ample. One might suppose, on theoretical grounds, that the constant change of process would entail some interference and thus make such tests more difficult than other forms covering the same examples. Myers and Myers have contributed some data to show that errors are more prevalent in mixed arrangements of examples, but they administered the tests orally and their data do not determine the comparative difficulty of mixed-fundamental and other types of tests.1 Lessenger found that mistakes due to misreading the words preceding the examples accounted for as much as six months loss in age scores on

Read the paper · More papers on PaperTik