500 



SELECTION AND TRAINING 



But the tank of oxygen was exhausted, and 

 only one man inflated his apparatus with air 

 and made the ascent. The others, unwilling 

 to make the ascent on air, even though it is 

 known to be practicable, remained in the 

 bell, were taken back into the sub, and two 

 of the three made no further attempts even 

 though they were certain to die in the sunken 

 boat. The third man directed solution of 

 the oxygen problem and led the next escape. 

 In one group, two officers were unable to 

 escape, one due to exhaustion and the other 

 due to both exhaustion and a minor injury. 

 The former remained in the ship and told the 

 men they could try to escape if they wished, 

 but he was through trying. The other 

 rested and escaped in a later group. Still 

 another case is that of the officer who, despite 

 wounds about the head, was self-possessed 

 enough to direct destruction of confidential 

 papers and later to make the ascent, but 

 loaded upon himself two pistols and a bayo- 

 net, presumably to do battle with the de- 

 stroyers topside. 



This revealing tragedy draws attention to 

 the great differences that lie beneath the 

 surface of men who are an excellent group of 

 brave and well-qualified submariners. The 

 more ordinary crises of undersea warfare 

 rarely reveal personality contrasts so 

 sharply, but it is probably true that under 

 depth-charging, crash dives, battle surface, 

 or other stress conditions, some men are 

 more disorganized than others. This is a 

 most difficult task order for prediction; not 

 that we need to predict which men will es- 

 cape under the rare conditions the Tang 

 experienced, but because we need to know 

 which men keep their heads in emergencies 

 or even outdo themselves, which men per- 

 form without slackening even when the odds 

 are heavily against them, which men take the 

 hazardous course rather than the cautious 

 pne, etc. 



,Screening Devices 



The first and greatest use of personality 

 tests is to screen out undesirable men. Such 



tests (or, rather, questionnaires) have been 

 widely used for screening of recruits, along 

 with psychiatric interviews. Any number of 

 studies have proved that such tests do iden- 

 tify most of the recruits who should be inter- 

 viewed with care (57, pp. 36-52). But such 

 tests are much less effective in dealing with 

 superior groups, such as submariners (3). 

 This does not show that personality is irrele- 

 vant; it only indicates that more powerful 

 tools are needed to identify excellent pros- 

 pects than to screen out poor ones. 



There are four basic approaches worth 

 consideration as ways of bettering person- 

 ality screening. One is the forced-choice. 

 Instead of asking the man "Can you be de- 

 pended upon — yes or no?", the test forces 

 the man to choose between two alternatives : 

 "I've got 'guts' " — "I can be depended 

 upon." Items of this type were used in the 

 Personal Inventory, long form, one of the 

 most effective tests to date. Such a test is 

 relatively free from the usual weakness that 

 men will not tell the truth about themselves. 



Personality tests, including forced-choice 

 devices, are much improved if an empirical 

 key is established. To develop such a key, 

 the test is given to a large number of men, 

 and ratings are later obtained to determine 

 which men worked out well and which pre- 

 sented personality problems. The papers 

 are analyzed to see what answers are given 

 most often by the "problem" cases. Any 

 new man who gives those answers is then 

 identified as a potential problem and can be 

 given special attention in processing. Jur- 

 gensen has applied the method in employee 

 selection (30) , using a technique quite appli- 

 cable to military situations. 



Projective tests are a third alternative re- 

 quiring further investigation. Such tests 

 present the man with an indefinite stimulus 

 and require him to react to it. He may say 

 what he sees in an inkblot, complete an un- 

 finished sentence, make up a story about a 

 dramatic picture, or the like. These tests 

 have been very effective in clinical work; 

 they evidently do tell a great deal about 



