CROSS-LAGGED AND SYNCHRONOUS COMMON FACTORS IN PANEL DATA (FROM STRUCTURAL EQUATION MODELS IN THE SOCIAL SCIENCES, 1973, BY ARTHUR S GOLDBERGER AND OTIS D DUNCAN - SEE NCJ-52004)

• NCJ Number: 52012
• Author(s): D A KENNY
• Date Published: 1973
• Length: 13 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

THE USE OF PANEL DATA FOR NONEXPERIMENTAL INFERENCE WITH CONTINUOUS VARIABLES, BASED ON THE TECHNIQUE OF CROSS-LAGGED PANEL CORRELATION, IS EXPLORED.

Abstract

TWO DIFFERENT TYPES OF FACTOR MODELS ARE CONSTRUCTED FOR EXPLAINING A CORRELATION BETWEEN TWO VARIABLES. IN THE FIRST MODEL, IT IS SPECIFIED THAT THE CORRELATION BETWEEN OBSERVABLES IS DUE TO A SYNCHRONOUS COMMON FACTOR ONE THAT CONTEMPORANEOUSLY AFFECTS BOTH OBSERVABLES. IN THE SECOND MODEL, IT IS STATED THAT THE CORRELATION BETWEEN OBSERVABLES IS DUE NOT ONLY TO A SYNCHRONOUS COMMON FACTOR BUT ALSO TO A COMMON FACTOR THAT AFFECTS THE FIRST VARIABLE CONTEMPORANEOUSLY AND AFFECTS THE SECOND VARIABLE AFTER SOME DELAY. THIS SECOND VARIABLE IS A CROSS-LAGGED COMMON VARIABLE. THE OBJECTIVE IS TO FIND RESTRICTIONS ON OBSERVABLE CORRELATIONS THAT SERVE TO DISCRIMINATE BETWEEN THE TWO MODELS. TWO-WAVE TWO-VARIABLE MODELS, THREE-WAVE TWO-VARIABLE MODELS, AND TWO-WAVE MULTIVARIATE MODELS ARE EMPLOYED TO TEST FOR A SYNCHRONOUS COMMON FACTOR IN PANEL DATA. SUPPORTING EQUATIONS ARE INCLUDED IN THE TEXT OF THE ARTICLE AND IN AN APPENDIX. (DEP)