U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

EXPLORATORY DATA ANALYSIS

NCJ Number
69184
Author(s)
F HARTWIG; B E DEARING
Date Published
1979
Length
83 pages
Annotation
EXPLORATORY DATA ANALYSIS (EDA) TECHNIQUES, APPROPRIATE FOR BOTH QUALITATIVE AND QUANTITATIVE DATA ANALYSIS, ARE PRESENTED.
Abstract
EDA TECHNIQUES ARE EMPLOYED IN BIVARIATE AND MULTIVARIATE ANALYSES IN THE SOCIAL SCIENCES, AND EMPHASIZE VISUAL DISPLAYS TO REVEAL VITAL INFORMATION ABOUT THE DATA BEING EXAMINED. CONSIDERABLE ATTENTION IS GIVEN TO TECHNIQUES FOR EXAMINING UNIVARIATE DISTRIBUTIONS OF DATA AND BIVARIATE RELATIONSHIPS BETWEEN VARIABLES. FOR UNIVARIATE DISTRIBUTIONS, THESE TWO VISUAL TECHNIQUES CAN YIELD VALUABLE INFORMATION ABOUT THE SHAPE OF A DISTRIBUTION. THE TECHNIQUES FOR EXAMINING RELATIONSHIPS BETWEEN TWO VARIABLES ARE DISCUSSED, AND THE SCATTER PLOT AND RESIDUAL PLOT TECHNIQUES ARE ILLUSTRATED. ILLUSTRATIONS OF VARIOUS TECHNIQUES ARE GIVEN USING SOCIOECONOMIC AND POLITICAL DATA FROM THE AMERICAN STATES. ONCE ELEMENTARY EDA TECHNIQUES HAVE BEEN USED TO EXAMINE THE SHAPE OF A DISTRIBUTION OR THE RELATIONSHIP BETWEEN TWO VARIABLES, MORE ADVANCED TECHNIQUES ARE AVAILABLE TO 'REEXPRESS' DATA TO OVERCOME PROBLEMS IDENTIFIED BY THE ELEMENTARY TECHNIQUES. THE IMPORTANCE OF REEXPRESSION IN DATA ANALYSIS AND SEVERAL MATHEMATICAL FUNCTIONS WHICH CAN BE USED TO REEXPRESS INTERVAL DATA ARE PRESENTED. A DISCUSSION OF HOW THE EDA PERSPECTIVE CAN BE USED IN MULTIVARIATE ANALYSIS CONCLUDES THE PAPER. THE RESIDUAL PLOT TECHNIQUE IS USED TO DECIDE WHICH ADDITIONAL VARIABLES SHOULD BE ADDED AT WHAT STEP IN A MULTIPLE REGRESSION ANALYSIS. TABLES, DIAGRAMS, AND A GLOSSARY ARE PROVIDED. FORTY REFERENCES ARE GIVEN. (AUTHOR ABSTRACT MODIFIED)