A Primer on Regression Artifacts

by Campbell, Donald T.; Kenny, David A.

• ISBN13: 9781572304826
• ISBN10: 1572304820
• Format: Hardcover
• Copyright: 1999-08-05
• Publisher: The Guilford Press

Summary

Regression toward the mean is a complex statistical principle that plays a crucial role in any research involving the measurement of change. This primer is designed to help researchers more fully understand this phenomenon and avoid common errors in interpretation. The book presents new methods of graphing regression toward the mean, facilitating comprehension with a wealth of figures and diagrams. Special attention is given to applications related to program or treatment evaluation. Numerous concrete examples illustrate the ways researchers all too often attribute effects to an intervention or other causal variable without considering regression artifacts as an alternative explanation for change. Also discussed are instances when problems are actually created, instead of solved, by "correction" for regression toward the mean. Throughout, the authors strive to use nontechnical language and to keep simulations and formulas as accessible as possible.

Author Biography

Donald T. Campbell, PhD, before his death in 1996 was University Professor of Social Relations, Psychology, and Education at Lehigh University. He had previously taught at Ohio State University, the University of Chicago, Northwestern University, and Syracuse University. He was a member of the National Academy of Sciences and a President of the American Psychological Association. He was the recipient of 9 honorary doctorates.

David A. Kenny, PhD, is Professor of Social Psychology at the University of Connecticut. He has been a visiting professor at Oxford University and Arizona State University, and was a Fellow at the Center for Advanced Study in the Behavioral Sciences.

Table of Contents

Graphical Introduction	1	(20)
From the Scatter Plot to the Correlation Coefficient	3	(9)
The Pair-Link and Galton Squeeze Diagrams	12	(4)
Backward Prediction	16	(1)
Biology and Regression toward the Mean	17	(1)
A Golden Oldie: McNemar's Illustration	18	(1)
Conclusion	18	(3)
Mathematical and Special Cases	21	(16)
Generalization to the Case with Unequal Means and Variances	21	(4)
Formal Definition and Formulas	25	(3)
Frequently Asked Questions about Regression toward the Mean	28	(7)
Conclusion	35	(2)
Regression Artifacts Due to Extreme Group Selection	37	(16)
Pre--Post Design and Regression Artifacts	37	(3)
Examples	40	(11)
Conclusion	51	(2)
Regression Artifacts Due to Matching	53	(15)
Detailed Illustration of the Limits of Matching	55	(10)
Direction of Bias	65	(1)
Conclusion	66	(2)
Regression Artifacts Due to Statistical ``Equating''	68	(19)
Look at the ``Treatment Effect'' on the Covariate	70	(1)
Illustration of the Bias in Statistical Equating	71	(3)
Direction of Bias	74	(4)
When Statistical Equating Works	78	(1)
Alternatives to Statistical Equating	79	(3)
Examples	82	(2)
Conclusion	84	(3)
Regression Artifacts in Change Scores	87	(14)
Correlations with Change Scores	88	(2)
Reliability of Change Scores	90	(1)
The Measurement of Change	91	(6)
Correlates of Change	97	(1)
Conclusion	98	(3)
Regression Artifacts in Time-Series Studies	101	(19)
The Connecticut Crackdown on Speeding	102	(3)
The Offset Effect of Psychotherapy	105	(5)
Time Series of HIV Treatment in Clinical Trials	110	(6)
Arima Modeling and Regression Artifacts	116	(2)
Conclusion	118	(2)
Regression Artifacts in Longitudinal Studies	120	(21)
Over-Time Correlational Structure	121	(10)
Regression Artifacts in Multiwave Studies	131	(8)
Conclusion	139	(2)
Cross-Lagged Panel Correlation Analysis	141	(14)
What Is CLPC Analysis?	141	(5)
Complications in CLPC Analysis	146	(2)
The Rogosa Critique	148	(3)
CLPC as a Special Case of the Multitrait--Multimethod Matrix Model	151	(2)
Conclusion: What Is to Be Done?	153	(2)
Conclusion	155	(18)
Common Sense and Data Analysis: A Brief Epistemological Exegesis	155	(3)
Time-Reversed Analysis	158	(5)
Graphical Display of Results	163	(1)
The Importance of Research Design	164	(1)
Careful Consideration of Plausible Rival Hypotheses	165	(2)
Regression and Prediction	167	(3)
Conclusion	170	(3)
Glossary of Terms	173	(6)
Glossary of Symbols	179	(2)
Appendix A. Dice-Rolling Program and Data Sets Used as Illustrations	181	(2)
Appendix B. The Computation of Autocorrelations	183	(2)
References	185	(10)
Index	195