| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
STATISTICAL CORNER |
From the Department of Psychology (N.J.S.C.), University of California, San Diego, La Jolla, California; Department of Psychiatry, Columbia University (R.P.S.), New York, New York; School of Sport & Exercise Sciences, University of Birmingham (D.C.), Birmingham, England; and the Department of Epidemiology, University of California Los Angeles (S.G.), Los Angeles, California.
Address correspondence and reprint requests to Nicholas Christenfeld, PhD, Department of Psychology, University of California, San Diego, La Jolla, CA 92093-0109. E-mail: nicko{at}ucsd.edu
ABSTRACT
ABSTRACT: When experimental designs are premature, impractical, or impossible, researchers must rely on statistical methods to adjust for potentially confounding effects. Such procedures, however, are quite fallible. We examine several errors that often follow the use of statistical adjustment. The first is inferring a factor is causal because it predicts an outcome even after "statistical control" for other factors. This inference is fallacious when (as usual) such control involves removing the linear contribution of imperfectly measured variables, or when some confounders remain unmeasured. The converse fallacy is inferring a factor is not causally important because its association with the outcome is attenuated or eliminated by the inclusion of covariates in the adjustment process. This attenuation may only reflect that the covariates treated as confounders are actually mediators (intermediates) and critical to the causal chain from the study factor to the study outcome. Other problems arise due to mismeasurement of the study factor or outcome, or because these study variables are only proxies for underlying constructs. Statistical adjustment serves a useful function, but it cannot transform observational studies into natural experiments, and involves far more subjective judgment than many users realize.
Key Words: confounds, • risk factors, • statistical control, • mediators, • covariates.
Abbreviations: BP = blood pressure;; SES = socioeconomic status;; MI = myocardial infarction.
This article has been cited by other articles:
 |
|
 |
|
  N. Agahi and M. G. Parker Leisure Activities and Mortality: Does Gender Matter? J Aging Health, October 1, 2008; 20(7): 855 - 871. [Abstract] [PDF]
|
|
|
|
 |
|
 A. Z. Fan, M. Russell, T. Naimi, Y. Li, Y. Liao, R. Jiles, and A. H. Mokdad Patterns of Alcohol Consumption and the Metabolic Syndrome J. Clin. Endocrinol. Metab., October 1, 2008; 93(10): 3833 - 3838. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Carroll, A. C. Phillips, and G. Der Body Mass Index, Abdominal Adiposity, Obesity, and Cardiovascular Reactions to Psychological Stress in a Large Community Sample Psychosom Med, July 1, 2008; 70(6): 653 - 660. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Korniewicz, T. Clark, and Y. David A National Online Survey on the Effectiveness of Clinical Alarms Am. J. Crit. Care., January 1, 2008; 17(1): 36 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Vandenbroucke, E. v. Elm, D. G. Altman, P. C. Gotzsche, C. D. Mulrow, S. J. Pocock, C. Poole, J. J. Schlesselman, M. Egger, and for the STROBE initiative Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration Ann Intern Med, October 16, 2007; 147(8): W-163 - W-194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Hayden, P. Cote, and C. Bombardier Evaluation of the quality of prognosis studies in systematic reviews. Ann Intern Med, March 21, 2006; 144(6): 427 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kim and B. M Popkin Commentary: Understanding the epidemiology of overweight and obesity--a real global public health concern Int. J. Epidemiol., February 1, 2006; 35(1): 60 - 67. [Full Text] [PDF]
|
|
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
