Sex And Personality Studies In Masculinity And Femininity by Terman Lewis M

Author:Terman,, Lewis M.
Language: eng
Format: azw
Publisher: Read Books Ltd.
Published: 2013-04-15T16:00:00+00:00

DERIVATION OF “I” SCORE WEIGHTS

The item comparisons could be made only for Form A of the M-F test, as Form B was not administered to the invert group. The techniques used in deriving the “I” scale were essentially the same as those by which response weights were derived in our preliminary experiments with the various parts of the M-F test, except that in this case two contrasting male groups were used instead of a male and a female group. That is, for each item of Form A of the M-F test the percentages of the two groups answering the item in a given way were computed. As each item has from two to four multiple-choice responses, percentages had to be computed for a total of some 1,560 possible responses. The differences between these percentages for the two groups were then translated into “I” score weights, a weight for each of the possible responses.

As in the case of the M-F test, the question arose whether to use unit or weighted scores; in other words, whether to make the weights uniformly +1 or −1, according as a given percentage is significantly higher for one group than for the other, or instead to give larger weights in proportion to the magnitude and reliability of the difference. It will be recalled that in the case of the M-F test weighted scores were finally abandoned in favor of unit scores. The latter reduce the labor of scoring, are about as reliable as weighted scores, and possibly more defensible psychologically in that they avoid overweighting large sex differences that may reflect chiefly the influences of differing environments. The world of men is in reality a very different world from that of women. In the present experiment, however, we are concerned with two male groups, and the psychological objection to the use of weighted scores is less cogent. Weights have therefore been employed. These range from +30 to −30, the plus weights being arbitrarily assigned to responses more frequently checked by the invert group. The weights for all the responses are reproduced in Appendix IV, so that anyone who wishes to use the “I” scale will be able to provide himself with a scoring key. Further investigation is desirable to determine whether a narrower range of weights, say from 1 to 3, would be better than the wide range used.2

In order to ascertain whether or not the “I” scores possess a peculiar significance of their own, that is, whether they are measuring something other than that measured by the M-F scores proper, all the test blanks which had been filled out by the invert group were rescored on the basis of the “I” weights. The distribution of the 82 scores3 resulting is shown in Table 63. As will be noted immediately, the total scores are practically all positive, and, on the whole, quite large. The range is almost two thousand points, which indicates wide individual differences among the members of the group and that the “I” scores are discriminative of these differences.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.