Multicollinearity may lead to artificial interaction

An example from a cross sectional study of biomarkers

Research output: Contribution to journalArticle

Abstract

Collinearity is the situation which arises in multiple regression when some or all of the explanatory variables are so highly correlated with one another that it becomes very difficult, if not impossible, to disentangle their influences and obtain a reasonably precise estimate of their effects. Suppressor variable is one of the extreme situations of collinearity that one variable can substantially increase the multiple correlation when combined with a variable that is only modestly correlated with the response variable. In this study, we describe the process by which we disentangled and discovered multicollinearity and its consequences, namely artificial interaction, using the data from cross-sectional quantification of several biomarkers. We showed how the collinearity between one biomarker (blood lead level) and another (urinary trans, trans-muconic acid) and their interaction (blood lead level * urinary trans, trans-muconic acid) can lead to the observed artificial interaction on the third biomarker (urinary 5-aminolevulinic acid).

Original languageEnglish (US)
Pages (from-to)404-409
Number of pages6
JournalSoutheast Asian Journal of Tropical Medicine and Public Health
Volume28
Issue number2
StatePublished - Jun 1997

Fingerprint

Cross-Sectional Studies
Biomarkers
Aminolevulinic Acid
Lead
muconic acid

ASJC Scopus subject areas

  • Medicine(all)

Cite this

@article{df7a7d6aabdb49b28f02060160d1b411,
title = "Multicollinearity may lead to artificial interaction: An example from a cross sectional study of biomarkers",
abstract = "Collinearity is the situation which arises in multiple regression when some or all of the explanatory variables are so highly correlated with one another that it becomes very difficult, if not impossible, to disentangle their influences and obtain a reasonably precise estimate of their effects. Suppressor variable is one of the extreme situations of collinearity that one variable can substantially increase the multiple correlation when combined with a variable that is only modestly correlated with the response variable. In this study, we describe the process by which we disentangled and discovered multicollinearity and its consequences, namely artificial interaction, using the data from cross-sectional quantification of several biomarkers. We showed how the collinearity between one biomarker (blood lead level) and another (urinary trans, trans-muconic acid) and their interaction (blood lead level * urinary trans, trans-muconic acid) can lead to the observed artificial interaction on the third biomarker (urinary 5-aminolevulinic acid).",
author = "Pornchai Sithisarankul and Weaver, {Virginia Marie} and Marie Diener-West and Strickland, {Paul Timothy}",
year = "1997",
month = "6",
language = "English (US)",
volume = "28",
pages = "404--409",
journal = "Southeast Asian Journal of Tropical Medicine and Public Health",
issn = "0125-1562",
publisher = "Southeast Asian Ministers of Education Organisation",
number = "2",

}

TY - JOUR

T1 - Multicollinearity may lead to artificial interaction

T2 - An example from a cross sectional study of biomarkers

AU - Sithisarankul, Pornchai

AU - Weaver, Virginia Marie

AU - Diener-West, Marie

AU - Strickland, Paul Timothy

PY - 1997/6

Y1 - 1997/6

N2 - Collinearity is the situation which arises in multiple regression when some or all of the explanatory variables are so highly correlated with one another that it becomes very difficult, if not impossible, to disentangle their influences and obtain a reasonably precise estimate of their effects. Suppressor variable is one of the extreme situations of collinearity that one variable can substantially increase the multiple correlation when combined with a variable that is only modestly correlated with the response variable. In this study, we describe the process by which we disentangled and discovered multicollinearity and its consequences, namely artificial interaction, using the data from cross-sectional quantification of several biomarkers. We showed how the collinearity between one biomarker (blood lead level) and another (urinary trans, trans-muconic acid) and their interaction (blood lead level * urinary trans, trans-muconic acid) can lead to the observed artificial interaction on the third biomarker (urinary 5-aminolevulinic acid).

AB - Collinearity is the situation which arises in multiple regression when some or all of the explanatory variables are so highly correlated with one another that it becomes very difficult, if not impossible, to disentangle their influences and obtain a reasonably precise estimate of their effects. Suppressor variable is one of the extreme situations of collinearity that one variable can substantially increase the multiple correlation when combined with a variable that is only modestly correlated with the response variable. In this study, we describe the process by which we disentangled and discovered multicollinearity and its consequences, namely artificial interaction, using the data from cross-sectional quantification of several biomarkers. We showed how the collinearity between one biomarker (blood lead level) and another (urinary trans, trans-muconic acid) and their interaction (blood lead level * urinary trans, trans-muconic acid) can lead to the observed artificial interaction on the third biomarker (urinary 5-aminolevulinic acid).

UR - http://www.scopus.com/inward/record.url?scp=0031151386&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031151386&partnerID=8YFLogxK

M3 - Article

VL - 28

SP - 404

EP - 409

JO - Southeast Asian Journal of Tropical Medicine and Public Health

JF - Southeast Asian Journal of Tropical Medicine and Public Health

SN - 0125-1562

IS - 2

ER -