Construct validity of a paper-and-pencil test of manual dexterity.

Abstract

Present tests of manual dexterity are invariably designed for use as individual or smallgroup tests which require relatively cumbersome equipment for their administration. Large scale preliminary assessment of manual skills, by means of a paper-and-pencil group test, would serve a purpose similar to the one for which paper-and-pencil tests of mechanical ability were devised (e.g., the Minnesota Paper Form Board, MacQuarrie Test of Mechanical Ability). The approach has been strongly supported by Harrel (1939) who suggested that the aptitudes measured by most apparatus tests might be measured just as successfully by properly constructed paper-and-pencil tests. The intent of this study was to devise and validate a paper-and-pencil test wherein dexterity might be measured by requiring manipulation of the writing instrument itself. Four paper-and-pencil subtests were used. Two of these were designated the Preferred ITand and Non-preferred Hand subtests which required S to grasp, alternately (using left or right hand as required) a red and then a blue pencil with which he placed an x in small squares arranged in columns on a sheet. Two other subtests, designated Two-hand and Assembly tests, required the use of both hands simultaneously with two-ended, dualcolored pencils that S rotated in each hand after each marking response. Fifty undergraduate Ss were given a battery of tests consisting of the Paper and Pencil Dexterity test ( 4 subtests), the Purdue Pegboard ( 4 subtests), the Pennsylvania Bimanual Work Sample ( 2 subtests) and two tests designed to measure manual speed ( a dotting test and a line drawing test). Construct validity was determined by factor analysis of the battery and examination of the resulting paper-and-pencil test loadings in the factor structure. Of the 66 intercorrelations in the 12 X 12 matrix, all were positive and 51 were significantly greater than zero. Four meaningful factors were extracted and rotated using a varimax solution. These were identified as General Manual Skill, Fine Finger Coordinarion, Manual Motor Speed and Gross Manual Manipulation. Although the Two-hand and Assembly subtests were somewhat more highly loaded on speed than would be desired, the four paper-and-pencil subtests generally fell into a pattern of loadings with the other tests that warrants determining their predictive validity against job-performance criteria.