the highly detailed objective data available on each aquanaut's reactions to 

 the undersea environment. 



DATA MANAGEMENT PROCEDURES 



John Wilhelm 

 The University of Texas at Austin 



Two major types of data were collected in the Virgin Islands. Observational 

 data were punched on site by University of Texas behavior observers onto IBM 

 portapunch cards. Other measures were collected before and during each mission 

 on questionnaires and mailed to Austin or California for keypunching. 



Each card punched was hand checked on site for obvious errors; mistakes were 

 often corrected before distance from the event made this impossible. Cards 

 were airmailed to Bellcomm, Inc. weekly. Bellcomm time- sequenced the cards, 

 stored reformated data on magnetic tape, and relayed to Texas the data as 

 punched IBM cards, formated printouts of the sequenced data, and card listings. 

 At the University of Texas, data were screened by computer to eliminate any 

 invalid data values and machine processing errors. Due to the large volume of 

 data generated during each mission, four preliminary data reduction programs 

 were written.* STATDAY gives daily totals on each variable for each diver 

 from Status Record; MEALDAY, COMMDAY, and DIVEDAY similarly combine data from 

 the Meal Record, Communication Record, and Dive Record to give daily summaries. 

 All data reduction programs compute values for either a complete day or for a 

 whole mission and print and punch these summaries for inspection and further 

 statistical analyses. 



The major preliminary analyses to date have been on the Aquanaut Status Record 

 using program STATREC. This program and several others were written before the 

 project began to enable tabulation of Status Record data as missions progressed. 

 All variables are totaled and converted to percentage scores to allow compari- 

 son across teams. From the various activity categories, composite criteria for 

 total leisure, total marine science, and total work are computed. 



The utilization of each compartment and section of each compartment by each 

 diver is calculated. Some communications analyses are also derived from Status 

 Record; for example, talk/listen to topside, talk/listen to other aquanauts, 

 and noncommunication. A detailed gregariousness matrix is also computed con- 

 sisting of the time spent by each diver talking to each of all possible groups 

 formed from the other aquanauts. 



The combination of variables recorded on the Aquanaut Status Record lends itself 

 to easy calculation of location-specific activities. Territoriality scores for 

 each diver in each section, and the sections most used by each team as a whole, 

 have been derived for leisure activities, for work activities, and for total 

 activities. 



* A listing of any program mentioned in this report may be obtained on request 

 from the authors . 



VIII-27 



