skip navigation

PUBLICATIONS

Register for Latest Research

Stay Informed
Register with NCJRS to receive NCJRS's biweekly e-newsletter JUSTINFO and additional periodic emails from NCJRS and the NCJRS federal sponsors that highlight the latest research published or sponsored by the Office of Justice Programs.

NCJRS Abstract

The document referenced below is part of the NCJRS Virtual Library collection. To conduct further searches of the collection, visit the Virtual Library. See the Obtain Documents page for direction on how to access resources online, via mail, through interlibrary loans, or in a local library.

 

NCJ Number: 52005 Find in a Library
Title: STRUCTURAL EQUATION MODELS - AN OVERVIEW (FROM STRUCTURAL EQUATION MODELS IN THE SOCIAL SCIENCES, 1973, BY ARTHUR S GOLDBERGER AND OTIS D DUNCAN - SEE NCJ-52004)
Author(s): A S GOLDBERGER
Corporate Author: Seminar Press, Inc
United States of America
Date Published: 1973
Page Count: 18
Sponsoring Agency: Russell Sage Foundation
New York, NY 10065
Seminar Press, Inc
New York, NY 10003
Type: Report (Study/Research)
Format: Document
Language: English
Country: United States of America
Annotation: THE USE OF STRUCTURAL EQUATION MODELS IN SOCIAL SCIENCE RESEARCH IS ADDRESSED, WITH PARTICULAR ATTENTION TO SITUATIONS IN WHICH CAUSAL THEORIES ARE AVAILABLE.
Abstract: IN A STRUCTURAL EQUATION MODEL, EACH EQUATION REPRESENTS A CAUSAL LINK RATHER THAN A MERE EMPIRICAL ASSOCIATION. IN A REGRESSION MODEL, ON THE OTHER HAND, EACH EQUATION REPRESENTS THE CONDITIONAL MEAN OF A DEPENDENT VARIABLE AS A FUNCTION OF EXPLANATORY VARIABLES. IT IS THIS DISTINCTION THAT MAKES CONVENTIONAL REGRESSION ANALYSIS AN INADEQUATE TOOL FOR ESTIMATING STRUCTURAL EQUATION MODELS. THE INADEQUACY OF LEAST SQUARES REGRESSION ANALYSIS IS DEMONSTRATED IN THREE CASES, INVOLVING UNOBSERVABLE VARIABLES (ERRORS OF MEASUREMENT), SIMULTANEITY (RECIPROCAL CAUSATION), AND OMITTED VARIABLES (INADEQUATE CONTROL). WITH REGARD TO STRUCTURAL PARAMETERS, THE ISSUE OF IDENTIFYING OBSERVABLE AND UNOBSERVABLE VARIABLES IS CRUCIAL. TRANSFORMATIONS OR ROTATIONS ARE WIDELY USED IN CONVENTIONAL EXPLORATORY FACTOR ANALYSIS, BUT THIS TYPE OF ANALYSIS MAY HAVE LIMITED APPEAL AS A BASIS FOR STRUCTURAL MODELING. WHEN THE MOMENTS OF OBSERVABLE VARIABLES PROVIDE MORE THAN ENOUGH INFORMATION TO DETERMINE STRUCTURAL PARAMETERS, A MODEL IS OVERIDENTIFIED. IN THIS CASE, THERE ARE RESTRICTIONS OR CONSTRAINTS ON THE OBSERVABLE MOMENTS. FOR ESTIMATING A CONVENTIONAL OVERIDENTIFIED SIMULTANEOUS EQUATION MODEL (NO MEASUREMENT ERROR), ECONOMISTS FREQUENTLY USE THE TWO-STAGE LEAST SQUARES PROCEDURE INITIATED BY THEIL. IN A VARIETY OF OVERIDENTIFIED SITUATIONS, EFFICIENT ESTIMATING PROCEDURES CAN BE VIEWED AS DEVICES FOR RECONCILING CONFLICTING MOMENT EQUATIONS. THE STRUCTURAL EQUATION APPROACH IS APPLICABLE TO ALL SOCIAL SCIENCE FIELDS. (DEP)
Index Term(s): Analysis; Evaluation; Evaluation techniques; Mathematical modeling; Research; Sociology; Theory
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
To cite this abstract, use the following link:
http://www.ncjrs.gov/App/publications/abstract.aspx?ID=52005

*A link to the full-text document is provided whenever possible. For documents not available online, a link to the publisher's website is provided. Tell us how you use the NCJRS Library and Abstracts Database - send us your feedback.