still more significant than sleep (p <.03) . Only solitary recreation was not 

 significant. In all cases, time spent working or in categories associated 

 with total work decreased from the first to the second half of the short mis- 

 sions, while time spent in total leisure pursuits or sleeping increased. 



Less work and more leisure during the second-half of the mission was also the 

 rule for long missions, but the differences are not so striking. Only 4 vari- 

 ables differed significantly: habitat maintenance and sleep were both highly 

 significant (p ^.001); total work (£ <C03), and direct marine science (£ <.05). 

 Neither total~leisure nor total marine science differed significantly. It 

 appears that the difference in total work is almost totally accounted for by 

 differences in time spent on habitat maintenance. Thus, it is not unreasonable 

 to claim that, at least with regard to first-half/last-half comparisons, the 

 long missions evidence greater patterns of stability than do the short missions. 



Totals, percentages, and means a,ll represent safe and familiar computational 

 ground. Analysis of variance and probabilistic statements of significance are 

 common enough tools for most social scientists. With time series data we are 

 on less familiar ground. Thus, for our next step in examining the data, we 

 regarded ourselves as well advised to revert to one of the most primitive yet 

 clearest methods of data presentation--graphic representation. Graphs were 

 prepared (by computer) which depicted the daily flow of each aquanaut's activ- 

 ity, A sample extracted from one of the graphs is included here (Figure 3). 



^^;:^^\\^W^\\^\^^?^?^^^$^\^N,^^^^K^^ 



l^\^W\^ 



2=H 



4= El 



mrnij 



m i 



EZD D C 



^^^•^ 



3 m 



3 D 



ti^^V\\\V^\H 



9 10 11 12 13 14 16 16 17 



TIME (HOUR OF DAY: 0:00" 24:00) 



20 21 22 23 24 



Figure 3 One Days Activity 

 VI II -48 



