The authors of several of the early standardized tests devoted considerable labor and care to the determination of supposedly accurate weights, usually on the basis of difficulty, for the separate elements composing their tests. Within a few years, however, a strong tendency to discontinue the use of such weights appeared. Studies by Douglass and Spencer 1 and others showed such high coefficients of correlation between scores based upon such weights and so-called "unweighted" or raw scores, that is, those based upon one point for each element, that they led to the conclusion that the considerable amount of extra work involved in deriving and using weights was unnecessary. In a recent article Corey 2 has presented some data which scarcely support this conclusion. He had each item of a new-type test in educational psychology weighted by six instructors in that subject. From the results he determined the correlations between raw scores and those computed according to these six series of weights, and also the effects of the different weightings upon letter marks into which the test scores were transmuted. Five of the six coefficients with the raw scores ranged from .82 to .88, the other being .96. Approximately one-fourth of the papers were given the same letter mark according to all seven scores, and another fourth each according to six, five, and four of the seven. The marks according to the six instructors varied from those based on raw scores in from twenty-two to forty-nine per cent of the cases.Because of the considerable disagreement between these results and others previously obtained, the present writer was led to make two studies along the same line. Before proceeding to present his own data, however, he wishes to offer one comment on Corey's investi-
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.