MIGRATION IS THE PRINCIPAL COMPONENT OF DEMOGRAPHIC CHANGES IN MOST STATES AND LOCAL AREAS, AND THE METHODS USED TO ESTIMATE AND PROJECT MIGRATION DESERVE SPECIAL ATTENTION. CONVENTIONAL ESTIMATES AND PROJECTION PROCEDURES FOCUS ON INDIVIDUAL MOVEMENT AND THE NUMBER OF MOVES THAT HAVE ALREADY OCCURRED. RECENT RESEARCH ON MIGRANT BEHAVIOR SUGGESTS, HOWEVER, THAT THE APPROPRIATE USE OF ANALYSIS IN THE MIGRATION DECISION IS THE HOUSEHOLD, NOT THE INDIVIDUAL, AND THAT GREATER ATTENTION SHOULD BE DEVOTED TO THE STUDY OF INTENTION TO MIGRATE RATHER THAN THE FACT OF MIGRATION. PROBIT ANALYSIS PERMITS A MORE RIGOROUS SPECIFICATION OF THE RELATION OF DEMOGRAPHIC, ECONOMIC, AND SOCIAL FACTORS TO THE INDIVIDUAL MIGRATION DECISION, AND THIS PAPER PRESENTS SUCH A MULTIVARIATE PROBIT MODEL FOCUSING ON HOUSEHOLD MIGRATION PLANS. THE PAPER ALSO PRESENTS THE EMPIRICAL FINDINGS FOR AN APPLICATION OF THE MODEL TO A 1976 SURVEY OF 3,610 METROPOLITIAN DELAWARE HOUSEHOLDS. THERE IS ALSO A DISCUSSION OF THE PROBLEMS AND PROCEDURES SURROUNDING THE CONSTRUCTION OF A PROBIT MODEL INCORPORATING AGGREGATE INFORMATION, AN APPROACH TO MODEL BUILDING CONSIDERED POTENTIALLY USEFUL FOR PLANNING AND POLICY RESEARCH. TABULAR DATA, A NOTE, AND REFERENCES ARE GIVEN. (AUTHOR ABSTRACT MODIFIED--MJW)