individual who is not only skilled in his own profession but eager to acquire 

 knowledge about other lines of endeavor should be the most effective type of 

 group member. This reasoning would suggest that the individual who is single- 

 mindedly dedicated to his particular professional aim would not make an effec- 

 tive member of a mixed crew. That sharing of diverse interests aided cohesive- 

 ness and compatibility was also demonstrated in debriefing of several scientist- 

 aquanauts who remarked that one of the high points of their experience was the 

 exchange of professional information with other scientists who had completely 

 different professional backgrounds. 



An important aspect of subsequent analyses will be an attempt to isolate back- 

 ground characteristics which can predict whether or not an individual will 

 enter into cross-professional exchange. This area appears to be of sufficient 

 importance to merit intensive investigation. 



Effects of Partial Crew Rotation 



In two missions, the engineer was replaced at the midpoint of the mission. The 

 effects of the rotation of one crew member on the behavior of the remaining 

 members are of both theoretical and practical importance. Because there are 

 large differences in behavior as a function of time in habitat (which will be 

 discussed later) the analysis of the effects of a crew change in the middle of 

 a 20-day mission must take into account the normal changes over time. As a 

 first analysis of these data, the behavior of the four scientist-aquanauts 

 before and after engineer change was compared with that of the four scientist- 

 aquanauts during comparable time periods in the four 20-day missions without 

 crew rotation. This comparison was done by unweighted means analysis of vari- 

 ance contrasting missions without rotation and missions with rotation (between) 

 and first-half of mission with second-half of mission (within) . In other words, 

 the first and second halves of missions with an engineer change (rotation was 

 at mission midpoint) were compared with the first and second halves of the four 

 20-day missions without an engineer change. The latter missions served as a 

 control for the former. In this analysis, behavioral effects of crew shift will 

 be reflected in a significant group (rotation or non-rotation) by trials (first- 

 half versus second-half) interaction.* Means and significance levels for activ- 

 ity variables are shown in Table 8. 



As can be seen in the table, changing a crew member did not significantly affect 

 the major performance criteria, total marine science and total work. There was 

 also no significant effect on sleep and total leisure. 



Social relations were significantly influenced by engineer change. This is 

 shown in the significant interactions for co-recreation, solitary recreation, 

 and gregariousness. As can be seen in Table 8, the percentage of time spent in 

 total leisure did not differ between rotated and unrotated missions over time 

 (the F-ratio is less than 1) . However, the types of leisure activities did 



* By a significant interaction we mean that the pattern of scores between the 

 first and second halves of the mission differs in the two groups. For 

 example, if solitary recreation increased after engineer change in rotated 

 crews but decreased in non-rotated crews, there would be an interaction 

 effect. 



VIII-43 



