NCJ Number
45896
Date Published
1977
Length
96 pages
Annotation
MULTIPLE REGRESSION ANAYSIS IS USED TO DETERMINE THE LEVEL OF THREAT REPRESENTED BY A TERRORIST INCIDENT TO PROVIDE THE DECISION MAKER WITH EARLY WARNING AND A PERSPECTIVE FOR ALLOCATING RESOURCES.
Abstract
THE 'INTERNATIONAL TERRORISM: ATTRIBUTES OF TERRORIST EVENTS' DATA FILE OF THE OFFICE OF POLITICAL RESEARCH OF THE CENTRAL INTELLIGENCE AGENCY (ITERATE FILE) WAS SURVEYED FOR THOSE ASPECTS OF TERRORISM WHICH POSE BOTH OBVIOUS AND LESS OBVIOUS THREATS. ANALYSIS OF VARIOUS FACTORS ISOLATED FROM THE ITERATE FILE AND INDEPENDENT ANALYSES OF THE THREATS BY EXPERTS SHOWED LITTLE CORRELATION. THE MAJOR PORTION OF THE PAPER SUGGESTS IMPROVEMENTS IN DATA COLLECTION BY ITERATE TO PROVIDE BETTER MATERIAL FOR ANALYSES. THESE INCLUDE: (1) INCREASING THE NUMBER OF SOURCES USED FOR THE FILE TO INCLUDE MORE NEWSPAPERS AND REPORT MATERIALS; (2) INCORPORATING A SET OF VARIABLES TO MEASURE AND REPRESENT THE LINKAGES WHICH EXIST BETWEEN THE TERRORIST GROUP INVOLVED IN THE INCIDENT AND OTHER TERRORIST GROUPS; (3) INCORPORATING A SET OF VARIABLES TO MEASURE AND REPRESENT THE LINKAGES BETWEEN THE INVOLVED TERRORIST GROUP AND OTHER INTERNATIONAL ACTORS; (4) INCORPORATING A SET OF VARIABLES TO MEASURE AND REPRESENT CULTURAL AND ETHNIC TRADITIONS AND CONSTRAINTS FELT BY THE TERRORIST GROUP WHICH COULD INFLUENCE ITS METHODS OF OPERATION; (5) MEASURING AND REPRESENTING A TERRORIST GROUP'S DEMANDS, PATTERNS, AND NEGOTIATING BEHAVIOR; AND (6) REPRESENTING RESULTS OF CONTENT ANALYSIS OF TERRORIST STATEMENTS. MODELS NEED TO BE DEVELOPED FOR INTEGRATING THE OPINIONS OF EXPERTS IN VARIOUS COMMUNITIES INTO A SCALE TO REPRESENT THE OVERALL THREAT PERCEPTIONS OF HIGH-LEVEL DECISION MAKERS, A FEEDBACK MECHANISM WHICH WOULD UPDATE THREAT SCALES IN CHANGING NATIONAL SECURITY SITUATIONS IS ALSO NEEDED. APPENDIXES GIVE DIMENSIONS OF SALIENCY OF ENEMY PERCEPTIONS, DESCRIPTIONS OF TERRORIST EVENTS ANALYZED, MAILING LISTS FOR DATA AND RESPONSE INFORMATION, A COPY OF THE QUESTIONNAIRE SENT TO EXPERTS, AND SUMMARIES OF THE MULTIPLE REGRESSION ANALYSES RUN. THE FACTORS PROVED SO INTERRELATED THAT NO SINGLE ONE COULD BE PULLED OUT AS SIGNALING MORE THREAT THAN ANY OTHER. (GLR)