Premium
CAUSALITY, MEASUREMENT ERROR AND MULTICOLLINEARITY IN EPIDEMIOLOGY
Author(s) -
ZIDEK JAMES V.,
WONG HUBERT,
LE NHU D.,
BURNETT RICK
Publication year - 1996
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/(sici)1099-095x(199607)7:4<441::aid-env226>3.0.co;2-v
Subject(s) - multicollinearity , spurious relationship , econometrics , causality (physics) , observational error , statistics , regression analysis , variable (mathematics) , phenomenon , regression , confounding , errors in variables models , linear regression , variables , computer science , mathematics , physics , quantum mechanics , mathematical analysis
This paper demonstrates that measurement error can conspire with multicollinearity among explanatory variables to mislead an investigator. A causal variable measured with error may be overlooked and its significance transferred to a surrogate. The latter's significance can then be entirely spurious, in that controlling it will not predictably change the response variable. In epidemiological research, such a response may be a health outcome. The phenomenon we study is demonstrated through simulation experiments involving non‐linear regression models. Also, the paper presents the measurement error problem in a theoretical setting. The paper concludes by echoing the familiar warning against making conclusions about causality from a multiple regression analysis, in this case because of the phenomenon presented in the paper.