LEADERSHIP IN RELATION TO GROUP EFFECTIVENESS 



475 



distinction, but the relationships imphed 

 need working through. The terms "posi- 

 tive" and "negative" have a morahstic ring 

 which suggests that they be abandoned in 

 favor of more precise definitions of the fac- 

 tors thought to be at work. And the possi- 

 bihty that the effects of such factors may 

 vary depending on the presence or absence 

 of other factors, as well as on the dimension 

 of morale studied, should be explored. 



Problems of Investigation 



From what has been said, it should be 

 plain that we are dealing with complex re- 

 lationships about which we know very little 

 in any precise sense. Virtually all of the 

 statements made above are more in the na- 

 ture of hypotheses than of established find- 

 ings. Certain general problems confronting 

 investigation might be noted. 



First of all, research is handicapped by 

 lack of an adequate concept of morale. The 

 studies cited have employed various aspects 

 of behavior as morale measures. While 

 these may all be in line with the popular 

 conception of morale, there is no assurance 

 that they represent the same thing psy- 

 chologically. The multiple dimensions of 

 morale which will doubtless be discovered 

 on inquiry will not react in the same manner 

 to most influences. 



Further problems concern methods of in- 

 vestigation.^ It is apparent that many, per- 

 haps all, of the conditions influencing morale 

 cannot be treated in isolation — that morale 

 is a function of interacting factors. For 

 example, it was suggested that the effects of 

 certain conditions might depend upon 

 whether morale in general was already high 

 or low, or whether due to the effects of 

 other factors these conditions were perceived 

 in one way rather than another. These con- 

 siderations would argue for the use of fac- 

 torial designs wherever possible in research, 



' Various of these problems are discussed in 

 greater detail by Kornhauser (60). Good general 

 discussions of methodology are found in mono- 

 graphs by Greenwood (39) and Chapin (18). 



and for the limiting of conclusions to the 

 range of conditions actually studied. 



Due to the difficulties involved in apply- 

 ing strict experimental procedures in many 

 practical situations, frequent use has been 

 made of ex post facto procedures. A case 

 in point is the Research Branch study of job 

 assignment in relation to job satisfaction 

 (45, discussed above). As in that case, the 

 possibility frequently exists that other fac- 

 tors related to the dependent variable may 

 have accounted for the grouping of sub- 

 jects with respect to the presumably in- 

 dependent variable, and doubts on this ques- 

 tion can often be resolved only by use of 

 conventional experimental procedures in- 

 volving prior design and randomization. In 

 many cases the latter type of procedure can 

 be employed, of course, and where this is 

 true, the greater certainty which it provides 

 may easily warrant the additional trouble 

 involved. 



Finally, it should perhaps be stressed again 

 that simply asking people what is good or 

 bad about their situation may indicate that 

 something is good or bad, but it will not 

 necessarily tell what it is. This phenom- 

 enon is so familiar to clinicians and so ob- 

 vious in the literature of industrial relations 

 as to need no emphasis, but morale in- 

 vestigators have sometimes shown a sur- 

 prising trust in the insight and diagnostic 

 abilities of their subjects. 



Leadership in Relation to Group 

 Effectiveness 



The Concept of Leadership 



Leadership has long been regarded as the 

 most important factor in group effectiveness, 

 and doubtless properly so, but a general con- 

 ception of leadership conducive to fruitful 

 research has been wanting. In many cases 

 it has simply been confused with pre- 

 eminence or with office-holding. In most 

 cases it has been defined as a function of cer- 

 tain personality traits residing in the in- 

 dividual designated as leader. 



