502 



SELECTION AND TRAINING 



pie, see 25) ; parallel work on military men is 

 essential. 



Studies of Leadership 



Another facet of the same problem is that 

 no one now knows what sort of men make the 

 best Naval officers. Moreover, no one has 

 demonstrated that the man who is best as a 

 junior officer makes the best commander or 

 the best staff officer. Studies of the person- 

 ality of effective and ineffective oflScers 

 would be a major help in selecting new 

 officers. Such a study should go much 

 deeper than a few questionnaires. A thor- 

 ough case study of several top-quality offi- 

 cers by a competent clinician would cast 

 much light on the effective military person- 

 ality. One approach to the problem is now 

 in use by the American Institute of Research, 

 studying Air Force officers and ONR scien- 

 tists. This valuable technique tries to iden- 

 tify the actions which distinguish good and 

 poor men. If this could be supplemented by 

 studies of personality structure which leads to 

 good or poor actions, one might be able to 

 identify potential commanders even among 

 seventeen-year-old recruits. 



While the majority of the problems of 

 leadership are discussed in another chapter, 

 one problem is essentially a selection prob- 

 lem. That is the question as to whether a 

 more efficient team can be assembled by 

 matching the men to form a compatible 

 group. It seems reasonable that every man 

 would work better with some superiors and 

 co-workers than others, but there are no 

 studies on this point. There have been a 

 few attempts to staff airplanes with groups of 

 men who chose each other, and the same 

 thing is done in an informal way in assem- 

 bling submarine crews. Research on what 

 sort of team brings out the best in each man 

 would no doubt provide suggestions for 

 placement of men and ofl&cers. 



Criteria of Personality 



A major problem in determining the extent 

 to which personality problems arise in under- 



sea warfare is the tradition against recording 

 personality failures in service records. It is 

 commendable for officers to avoid condemn- 

 ing a man for a crackup; a face-saving 

 transfer or disqualification for medical rea- 

 sons is a sound administrative practice. 

 But some reliable method is needed to de- 

 termine the incidence of psychiatric prob- 

 lems. 



Since men passed by screening psychia- 

 trists develop anxieties and tensions in the 

 course of patrols, some method is needed to 

 monitor personality in the field. Probably 

 a procedure should be developed for a short 

 psychiatric interview with every member of a 

 submarine crew after every war patrol. It 

 is unsound to assume that initial screening 

 is a sufficient safeguard. Such further 

 screening would identify problems about to 

 come to the surface in erratic behavior and 

 permit men to receive rest when they need it. 



The Criterion 



One of the greatest weaknesses of selection 

 and training in the late war was the inade- 

 quacy of criteria. Without a suitable meas- 

 ure of how well men perform on the job, no 

 one can say what tests are permitting good 

 selection, which training methods are effec- 

 tive, and which men should be selected for 

 added responsibility. Nearly every research 

 report on a wartime selection program in- 

 cludes a comment that the research is in- 

 adequate because no fair criterion could be 

 obtained. 



A report from BuPers well demonstrates 

 the seriousness of criterion errors. When 

 poor tests were given to men graduating from 

 Basic Engineering School, it seemed that 

 prediction was poor, but that the Arithmetic 

 test was the best test for prediction. When 

 new and better achievement tests were de- 

 signed and given to graduates, not only was 

 validity raised, but it was found that a 

 different set of tests (especially Mechanical 

 Aptitude) selected the best men. The poor 

 criterion, in other words, led to recommenda- 

 tion of the wrong selection battery (44, pp. 



