TESTING MEN ASSEMBLERS 73 clip; the fourth a bicycle bell, etc. The task of the sub ject is to take the object in each compartment and as semble it. Twenty minutes were given for this test and Ae results recorded according to the scale of values worked °ut by Professor Stenquist. The object of this test was to discover mechanical ability. This experiment was conducted in two sections. In the first section all but the last of the tests named were given to thirty-one men engaged in action assembling. the second section of the experiment all the tests were g ly en to fifty-one finishing assemblers and twenty-six action assemblers. During the course of the first part of the experiment, all the assemblers were on a day-work father than a piece-work basis, and for this reason it was lrtl possible to obtain an objective rating of the men based °u their actual production. Therefore, in lieu of some- ning better, each foreman was asked to rank the men. t was carefully explained to him that his ranking should ^ based upon his opinion as to the possible production of men if they were allowed to work on a piece-work a sis. The fact that during a previous drive on production jfi°st of these men had been on piece-work made it possible r the foreman to make such a ranking on something more . an merely hypothetical grounds. The foremen’s rank- lr fgs were then mathematically compared with the ranking ° tbe men in the various tests, with the following results: Correlations No. 31—Spatial No. 34—3 Trials No. 31—Hand 1 Action Perception Dynamometer Sse mblers. . •56 .18 .29