METHODS CF INVESTIGATION 



The methods of investigation used 

 in this study are depicted in Figure 2. 

 Before the information search began, 

 goals were established and a conceptual 

 outline was created. A Procedures Man- 

 ual was prepared which contained steps 

 for information extraction and data 

 sheet completion, as well as a conceptu- 

 al outline of the project. Readers were 

 subsequently trained, using the Proce- 

 dures Manual, and their ability to ex- 

 tract relevant information was tested. 



Concurrent with the training pro- 

 cedure, the search for information was 

 begun using commercial search services, 

 primarily through the Oceanic and At- 

 mospheric Scientific Information System 

 (OASIS). This system is quite inclu- 

 sive of the commercial data bases that 

 may contain information regarding minor 

 shoreline structures. Approximately 

 220 of the 555 total information sources 

 were found using the OASIS search. 

 The balance of the information sources 

 came from bibliographies contained in 

 identified sources, questionnaires, lib- 

 raries, and interviews. 



Approximately 300 questionnaires 

 were distributed to institutions, govern- 

 ment agencies, and individuals that 

 might have relevant information. A 

 conceptual outline of the project accom- 

 panied the questionnaires. Monrespond- 

 ents and those respondents whose an- 

 swers needed clarification were contact- 

 ed by telephone. Where desirable, inter- 

 views and/or telephone calls were made 

 with persons supplying valuable infor- 

 mation in the questionnaire responses. 

 Approximately 40 interviews were con- 

 ducted. The map in Figure 3 shows the 

 areas of the United States where ques- 

 tionnaires were sent and where inter- 

 views were conducted. Materials were 

 accepted and entered into the data base 

 until the eighth month, at which time 

 searching was halted to facilitate timely 

 completion of the contract report. In- 

 formation was extracted from the litera- 

 ture and entered on data sheets by 

 structure type and region. The informa- 

 tion categories are contained in Figure 

 4. A bibliographic data sheet was also 

 completed for each source. That sheet 

 contained title, author, abstract, other 



pertinent citation information, key words, 

 and a rating. Articles were rated ac- 

 cording to their applicability and useful- 

 ness to the objectives of the project 

 (not for their scientific excellence or 

 validity) on a scale of one (excellent) 

 to five (poor). Those articles that were 

 reviewed, but not considered directly 

 applicable to the objectives of the study 

 were abstracted, but not keyworded. 



The two types of data sheets were 

 reviewed for punctuation and spelling 

 before being sent to keypunch. After- 

 keypunching, the data were again 

 checked for punctuation, spelling, con- 

 tents, and keypunching errors. The 

 data were then entered into the data 

 base (Figure 4). 



The data base management system 

 used by Eeak Consultants Incorporated 

 was System 2000 developed by MRI 

 Systems Corporation of Austin, Texas. 

 This system was made available through 

 Computer Sciences Corporation's (CSC) 

 INF0NET timesharing system on the 

 UNIVAC 1108 computer. In addition to 

 System 2000, Beak Consultants Incorpo- 

 rated used its own proprietary FOR- 

 TRAN programs to load the data base 

 and provide the formatted outputs. 

 System 2000 was used because of its 

 ability to handle the large amount of 

 data that were extracted from the infor- 

 mation sources and also because of the 

 multilevel on-line access capability 

 which provided assistance during the 

 report writing phase. 



The outputs produced from the 

 data base were an annotated bibliogra- 

 phy, keyword index, primary author 

 reference number index, and a printout 

 of information extracted. The informa- 

 tion was printed in a Region-Structure 

 hierarchy although System 2000 has the 

 capability of supplying a printout in 

 numerous hierarchies. Data base inter- 

 rogation during report writing was done 

 through various data base entry points. 

 Using the data outputs the existing in- 

 formation was evaluated and interpret- 

 ed. A text was written according to 

 the structure types and information cat- 

 egories presented previously in this 

 section. An evaluation of the existing 



