IDENTIFICATION, PARAMETER ESTIMATION, AND HYPOTHESIS TESTING IN RECURSIVE SOCIOLOGICAL MODELS (FROM STRUCTURAL EQUATION MODELS IN THE SOCIAL SCIENCES, 1973, BY ARTHUR S GOLDBERGER AND OTIS D DUNCAN - SEE NCJ-52004)

• NCJ Number: 52006
• Author(s): K C LAND
• Date Published: 1973
• Length: 31 pages
• Sponsoring Agency: Russell Sage Foundation
• Corporate Author: Seminar Press, Inc
• Publication Format: Document
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America
• Note: REVISION OF A PAPER PRESENTED AT A CONFERENCE COSPONSORED BY THE SOCIAL SCIENCE RESEARCH COUNCIL AND THE SOCIAL SYSTEMS RESEARCH INSTITUTE OF THE UNIVERSITY OF WISCONSIN, MADISON, NOV 12-16, 1970

Annotation

STANDARD THEORIES OF IDENTIFICATION, PARAMETER ESTIMATION, AND HYPOTHESIS TESTING ARE SPECIALIZED TO THE CLASS OF RECURSIVE MODELS WITH INDEPENDENT DISTURBANCES WHICH HAVE BEEN WIDELY APPLIED IN SOCIOLOGY.

Abstract

CERTAIN GENERAL FEATURES CHARACTERIZE TYPICAL APPLICATIONS OF SIMULTANEOUS EQUATION MODELS IN SOCIOLOGY: (1) MODELS USUALLY CONSIST OF A SET OF EQUATIONS THAT POSSESS A RANDOM DISTURBANCE OR ERROR TERM TO SUMMARIZE THE INFLUENCE OF UNMEASURED OR UNKNOWN VARIABLES ON THE MECHANISM OF INTEREST; (2) VARIABLES OF SOCIOLOGICAL MODELS ARE TYPICALLY NOT FUNCTIONS OF TIME; (3) MOST MODELS RULE OUT TWO-WAY CAUSATION AMONG SYSTEM VARIABLES; (4) MOST SIMULTANEOUS EQUATION MODELS ARE LINEAR IN VARIABLES AND DISTURBANCES; (5) MOST SOCIOLOGICAL MODELS IMPOSE THE ASSUMPTION THAT DISTURBANCE TERMS ARE CONTEMPORANEOUSLY INDEPENDENT; AND (6) IT IS ASSUMED GENERALLY THAT SYSTEM VARIABLES ARE MEASURED WITHOUT ERROR. ISSUES OF CONCERN IN SIMULTANEOUS EQUATION MODELS ARE IDENTIFICATION, PARAMETER ESTIMATION, AND HYPOTHESIS TESTING. SUPPORTING EQUATIONS AND EXAMPLES THAT ILLUSTRATE SUBSTANTIVE APPLICATIONS OF SIMULTANEOUS EQUATION MODELS ARE GIVEN. (DEP)