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.

    116/172 (page 104)
    Results of Addition Test (cont). Lrronp Coefficient of Variation Eeliability coefficient Eeliability coefficient Speed Accuracy Speed Acc. Speed Acc. I 28 + 1-8 19 + 1-2 -82 •33 •90 •50 II 24-6+ 2-0 22-4 + 1-8 •69 •46 •82 •63 III 27 + 2-2 8+ -6 -68 [0] •81 *IV 30 + 2-9 9-6+ -86 •93 •29 •96 •45 *V(a) 32 + 2-8 6 ±-5 -95 •22 •97 •36 Y{b) •98 •59 ■99 •74 (5) Bisection and Trisection of Lines. Every test paper contained ten printed lines, each 8 cm. long, for bisection; and ten printed lines, each 9 cm. long, for trisection. The lines were printed three in a row, and those situated immediately under others were shifted a little to one side. It was very certain, however, that the bisection or trisection of any one of the lines was influenced by the positions of the neighbouring lines and of the edges of the paper. This fact diminishes the value of the test, but does not of itself deprive the test of all use as a measure of one form of sensory discrimination. A more serious drawback was found to be the very great individual variability displayed, which made the reliability coefficients very low. In order to get a more reliable measure for Group V the results for bisection and trisection in both tests (i.e. of the division of 40 lines in all by each individual) were thrown together and the total taken as a measure of sensory discrimination. Onli/ one point of trisection was asked for, this being put in alternately towards the left and the right ends of the successive lines. Trisection was done very unsatisfactorily by the school-children, but, in the case of adults, gave a higher measure of reliability than did bisection. * One sum of 10 digits taken as unit; in other cases, the number of digits added was taken as the measure.