Even in this case the presupposition ought to be that the average characteristics found out with scientific exactitude by statistical and experimental methods, and not that they are simply deduced from superficial impressions.  I have found that just this race psychological diagnosis is frequently made in factories with great superficiality.  Some of the American industrial centres offer extremely favorable conditions for the comparative study of nationality.  I have visited many manufacturing establishments in which almost all workers are immigrants from foreign countries and in which up to twenty different nationalities are represented.  The employment officers there easily develop some psychological theories on the basis of which they are convinced that they are selecting the men with especial skill, knowing for each in which department he will be most successful.  They consider it settled that for a particular kind of activity the Italians are the best, and for another, the Irish, and for a third, the Hungarians, and for a fourth, the Russian Jews.  But as soon as these factory secrets have been revealed, you may be surprised to find that in the next factory a decidedly different classification of the wage-earners is in force.  In a gigantic manufacturing concern, I received the definite information that the Swedish laborers are preferable wherever a steady eye is needed, and in another large factory on the same street I was assured that just the Swedes are unfit for such work.  Sometimes this diversity of opinion is the result of different points of view.  In one factory in which a certain industrial operation is rather dangerous, they told me that they took no southern Europeans, especially no Italians and Greeks, because they are too hasty and careless in their movements, while they gladly filled the places with Irishmen.  In a quite similar factory, on the other hand, they had a prejudice against the Irishmen alone for this work, because the Irish laborers are too willing to run a risk and to expose themselves to danger.  Probably both psychological observations are on the whole correct, but in the first factory only the one and in the second factory only the other was recognized.  Much more thorough statistical inquiries than those which as yet exist, especially as to the actual differences of wages and piecework for wage-earners of various nationalities, would have to furnish a basis for such race psychological statements, until the time arrives when the psychological experiment comes to its own.

In a similar way so far we have to rely on general theories of group psychology when the psychological differences of the sexes are to be reckoned with in economic interests.  So long as laboratory methods for individual tests are not usual, the mental analysis of the general groups of men and women must form the background for industrial decisions.  To be sure, it is not difficult to emphasize certain mental traits as characteristic of women in general in contrast to