STATISTICS AND SOFTWARE FOR COMPLEX SAMPLE SURVEYS

LARGE-SCALE SURVEYS USING COMPLEX SAMPLE DESIGNS ARE FREQUENTLY CARRIED OUT BY GOVERNMENT AGENCIES CONCERNED WITHHEALTH ISSUES SUCH AS ALCOHOL USE AND RELATED PROBLEMS. THESTATISTICAL ANALYSIS TECHNOLOGY AVAILABLE FOR SUCH DATA IS BOTH LIMITED AND DIFFICULT TO ACCESS. THIS PROJECT CONCERNSTHE INVESTIGATION AND FURTHER DEVELOPMENT OF UNIVARIATE AND MULTIVARIATE STATISTICAL METHODS AND SOFTWARE FOR THE ANALYSIS OF DATA FROM COMPLEX SAMPLE DESIGNS. SPECIFICALLY, EXISTING STATISTICAL METHODS AND SOFTWARE WILLBE IDENTIFIED. IN PHASE I, IDEAS ABOUT DESIRED FEATURES OF PERSONAL COMPUTER AND MAINFRAME SOFTWARE WILL BE GENERATED AND THE POSSIBILITY OF THEIR PRACTICAL IMPLEMENTATION WILL BE INVESTIGATED. PARTICULAR ATTENTION WILL BE PAID TO THE POTENTIAL OF MICROCOMPUTER IMPLEMENTATION, BOTH IN TERMS OF COMPUTATIONAL CAPACITY AND USER-FRIENDLY INTERFACE. PHASE II WORK WILL IDENTIFY AVENUES FOR THE DEVELOPMENT OF MORE COMPLEX MULTIVARIATE STATISTICAL METHODOLOGY AND SOFTWARE TO BE CARRIED OUT WITH A VIEW TOWARD EXPANDING TODAY'S ANALYSIS OPTIONS WITH COMPLEX SAMPLES, INCLUDING OPTIONS UNIQUE TO MAINFRAME COMPUTERS. BOTH MICRO- AND MAINFRAME TECHNOLOGY OF THIS KIND SHOULD HAVE WIDE APPLICABILITY AND COMMERCIAL POTENTIAL GIVEN THE LARGE NUMBER OF USERS OF LARGE-SCALE SURVEY DATA.