NCJ Number
57692
Journal
Sociological Methods and Research Volume: 7 Issue: 4 Dated: SPECIAL ISSUE (MAY 1979) Pages: 451-474
Date Published
1979
Length
24 pages
Annotation
THE PROBLEMS ENCOUNTERED IN AND SEVERAL SOLUTIONS TO OVERCOME ANALYSIS PROCEDURES AND INFERENCES WHEN RATIOS ARE COMPUTED BY USING THE SAME OR HIGHLY CORRELATED DENOMINATORS ARE DISCUSSED.
Abstract
SOCIOLOGICAL AND DEMOGRAPHIC RESEARCH OFTEN USE VARIABLES COMPUTED AS RATIOS. WHEN THE DENOMINATORS ARE HIGHLY CORRELATED, OR IDENTICAL, AND WHEN THE RATIOS ARE USED IN CORRELATION, REGRESSION, AND PATH ANALYSIS, THEN CALCULATED COEFFICIENTS ARE INFLUENCED BY THE COMMON ELEMENTS IN THE DENOMINATORS AND THIS CAN LEAD TO SERIOUS MISINTERPRETATIONS. THIS PROBLEM MAY BE PRESENT IN EITHER CROSS-SECTIONAL OR LONGITUDINAL RESEARCH, AND IS SHOWN TO EXPAND FROM BIVARIATE CORRELATION AND REGRESSION TO PARTIAL CORRELATION AND MULTIPLE REGRESSION. ADVANTAGES AND DISADVANTAGES OF THREE SELECTED ALTERNATIVE CHANGE MODELS (DIFFERENCE SCORES, PROPORTIONS OR PERCENTAGES, AND RESIDUALIZED DIFFERENCE SCORES) ARE EXAMINED, WITH A FOCUS ON PATH ANALYSIS AND THE PROBLEM OF CORRELATED DENOMINATORS IN LONGITUDINAL PATH ANALYSES. IT IS SUGGESTED THAT, WHEN AN IDENTICAL DENOMINATOR EXISTS, IT CAN BE USED AS AN INDEPENDENT CONTROL VARIABLE IN STANDARD LEAST-SQUARES REGRESSION EQUATIONS CONSTRUCTED FROM THE NUMERATORS. WHEN, HOWEVER, THE DENOMINATORS ARE HIGHLY CORRELATED BUT NOT IDENTICAL, AS IS FOUND IN MOST CROSS-SECTIONAL RESEARCH AND IS VIRTUALLY INESCAPABLE IN LONGITUDINAL RESEARCH, THE USE OF RESIDUAL ANALYSIS CAN SOLVE THE PROBLEM OF CORRELATED DENOMINATORS. THE TEXT IS ILLUSTRATED WITH DIAGRAMS OF THE EQUATIONS USED. NOTES AND REFERENCES ARE PROVIDED. (AUTHOR ABSTRACT MODIFIED--MHP)