The essentials of mental measurement / by William Brown.

  • Brown, William, 1811-
Date:
1911
Catalogue details

Licence: In copyright

Credit: The essentials of mental measurement / by William Brown. Source: Wellcome Collection.

Provider: This material has been provided by King’s College London. The original may be consulted at King’s College London.

    86/172 (page 74)
    symmetrical function of p, this is not a sound course to adopt. In fact, if the numbers are small, reversal of one of the series may yet leave R negative. Pearson's own formula (e) is free from this drawback*. An important argument brought forward by Spearman f is one to the effect that absolute measurements are often hetero- geneous and that ranking removes this heterogeneity. There is an assumption underlying this argument which would seem to be of somewhat doubtful validity. All, or almost all, so-called mental measurement is measurement of the object. It is physical, not mental at all. The physical objects, the physical results can be measured—in terms of space and time—and they can be correlated. Psychical measurement has perhaps some plausibility in the case of sensation intensities, in the form of the sense-distances of Delboeuf; but this furnishes small justification for the claims made above on behalf of ranks. The unit of rank means nothing until the form of distribution is fixed, and this can hardly be determined on a purely psychical basis; at least it is difficult at present to conceive of even an attempt at such a determination. Another important passage occurs on p. 99 of the previously- quoted article, In this simplified form the standard or r method To .1 .V, f 1 1 ^S{v,-v,r Spearman apparently means the tormula 1 - -\T/]ai_-i\ ' shows itself to be solely distinguished from our short or R method by the fact that the differences of rank are squared. The efifect of squaring is to give more ' weight' to the extreme differences as compared with the medium ones. This is probably a considerable advantage in most physical measure- ments. But in other fields of research, and perhaps above all in psychology, these extreme cases are just the ones of most * Spearman endeavours to meet some of these criticisms in a recent paper; British Journal of Psychology, Vol. in. Pt. 3, Oct. lUlO. t C. Spearman: 'Foot-rule' for Measuring Correlation, Bntish Journal of Psychology, Vol. n. Pt. i. July 1906, p. 93. which is p, not the standard