468 



MORALE AND LEADERSHIP 



are reported on the basis of preliminary 

 trials to tell much concerning the functioning 

 of the group."* Another potentially valuable 

 projective device applicable to small groups 

 is the sociodramatic technique of Moreno 

 (77). 



Group Indices 



Several wTiters (1, 13) have called atten- 

 tion to various kinds of indices that might 

 be obtained from group records covering a 

 period of time. In a military situation these 

 might include statistics on disciplinary of- 

 fenses, men on "sick call," psychoneurotic 

 and psychosomatic disorders, venereal dis- 

 eases or other diseases such as malaria in 

 which incidence may depend on the taking of 

 routine precautions by the men, requests for 

 transfer, men striking for higher rates, and 

 so on. There are obvious problems in the 

 use of such indices: they can give only a kind 

 of average picture for a period of time, in 

 practice the necessary records may be incom- 

 plete or unavailable, and they may be influ- 

 enced unduly by extraneous factors. Nev- 

 ertheless, further exploration in this direction 

 may be worthwhile. 



Situation Tests 



Additional evidence of a more dynamic 

 sort might be gained by observing group 

 reactions in standardized stress situations. 

 In effect, whole groups might be subjected 

 to situation tests analogous to those recently 

 introduced in studies of individuals as poten- 

 tial leaders. Some suggestive possibilities 

 may be found in the literature, e.g., a simu- 

 lated fire with a group in a locked room (32), 

 intmsion of a critical stranger (66), false 

 news announcements (78), together with 

 various situations of a more conventional 

 sort, such as group problems, inter-group 

 competitions, etc. Measures of aggression, 

 fear, persistence, acceptance of destructive 

 rumors, etc., in such situations might prove 

 very revealing of a group's potentialities. 



^ Reported in The TAT Newsletter, 1948, 1, 

 ^o. 3, edited by R. R. Holt, Menninger Founda- 

 tion, Topeka, Kansas. 



While procedures of this sort would perhaps 

 have their greatest usefulness in relatively 

 small scale research investigations, it is not 

 unlikely that some realistic test situations 

 could be devised for more or less routine 

 application in conjunction with regular 

 group training programs, e.g., during shake- 

 do^vn, or that pertinent data might be gath- 

 ered in stress situations which already 

 occupy a place in such training. 



The Validity of Morale Measures 



It was pointed out earlier that direct meas- 

 ures of the success of groups in attaining 

 their goals may be influenced heavily by ad- 

 ventitious factors, and that our problem 

 concerns the relative effectiveness of differ- 

 ent groups when differences in these factors 

 are, at least conceptually, eliminated. Ac- 

 tually to eliminate such influences is ex- 

 tremely difficult (13), and for that matter, 

 relevant direct measures may in practice be 

 unobtainable. How then can it be shown, 

 as we have also demanded, that potential 

 measures of morale or group effectiveness are 

 valid? 



The answer, it would seem, is that we must 

 in some degree lift ourselves by our boot 

 straps. Certain potential measures have a 

 high degree of face validity. If a group 

 spends most of its time in wrangling or in 

 loafing, or has a high proportion of desert- 

 ers, it obviously cannot be functioning very 

 effectively. Whether or not such measures 

 are available, an analysis of interrelation- 

 ships among measures may also yield 

 relevant evidence. If several morale indi- 

 cators are hypothesized on the basis of re- 

 lated fundamental assumptions, and, when 

 measured, are found to be highly intercor- 

 related, this should justify at least a tenta- 

 tive presumption of validity. Likewise, if 

 certain effects on morale are predicted to 

 follow certain events, and a presumed morale 

 indicator shows the predicted effects, this 

 may be considered relevant evidence. 



Further evidence on validity may be ob- 

 tained in experimental group situations af- 

 fording control of extraneous influences on 



