Path analysis in genetic epidemiology: A critique

S. Karlin, E. C. Cameron, Ranajit Chakraborty

Research output: Contribution to journalReview article

25 Citations (Scopus)

Abstract

Path analysis, a form of general linear structural equation models, is used in studies of human genetics data to discern genetic, environmental, and cultural factors contributing to familial resemblance. It postulates a set of linear and additive parametric relationships between phenotypes and genetic and cultural variables and then essentially uses the assumption of multivariate normality to estimate and perform tests of hypothesis on parameters. Such an approach has been advocated for the analysis of genetic epidemiological data by D.C. Rao, N. Morton, C.R. Cloninger, L.J. Eaves, and W.E. Nance, among others. This paper reviews and evaluates the formulations, assumptions, methodological procedures, interpretations, and applications of path analysis. To give perspective, we begin with a discussion of path analysis as it occurs in the form of general linear causal models in several disciplines of the social sciences. Several specific path analysis models applied to lipoprotein concentrations, IQ, and twin data are then reviewed to keep the presentation self-contained. The bulk of the critical discussion that follows is directed toward the following four facets of path analysis: (1) coherence of model specification and applicability to data; (2) plausibility of modeling assumptions; (3) interpretability and utility of the model; and (4) validity of statistical and computational procedures. In the concluding section, a brief discussion of the problem of appropriate model selection is presented, followed by a number of suggestions of essentially model-free alternative methods of use in the treatment of complex structured data such as occurs in genetic epidemiology.

Original languageEnglish
Pages (from-to)695-732
Number of pages38
JournalAmerican Journal of Human Genetics
Volume35
Issue number4
StatePublished - 1 Jan 1983

Fingerprint

Molecular Epidemiology
Social Sciences
Structural Models
Medical Genetics
Lipoproteins
Linear Models
Phenotype

Cite this

Karlin, S., Cameron, E. C., & Chakraborty, R. (1983). Path analysis in genetic epidemiology: A critique. American Journal of Human Genetics, 35(4), 695-732.
Karlin, S. ; Cameron, E. C. ; Chakraborty, Ranajit. / Path analysis in genetic epidemiology : A critique. In: American Journal of Human Genetics. 1983 ; Vol. 35, No. 4. pp. 695-732.
@article{2bb2ee93f1dc465aa6693f8c6f0895c0,
title = "Path analysis in genetic epidemiology: A critique",
abstract = "Path analysis, a form of general linear structural equation models, is used in studies of human genetics data to discern genetic, environmental, and cultural factors contributing to familial resemblance. It postulates a set of linear and additive parametric relationships between phenotypes and genetic and cultural variables and then essentially uses the assumption of multivariate normality to estimate and perform tests of hypothesis on parameters. Such an approach has been advocated for the analysis of genetic epidemiological data by D.C. Rao, N. Morton, C.R. Cloninger, L.J. Eaves, and W.E. Nance, among others. This paper reviews and evaluates the formulations, assumptions, methodological procedures, interpretations, and applications of path analysis. To give perspective, we begin with a discussion of path analysis as it occurs in the form of general linear causal models in several disciplines of the social sciences. Several specific path analysis models applied to lipoprotein concentrations, IQ, and twin data are then reviewed to keep the presentation self-contained. The bulk of the critical discussion that follows is directed toward the following four facets of path analysis: (1) coherence of model specification and applicability to data; (2) plausibility of modeling assumptions; (3) interpretability and utility of the model; and (4) validity of statistical and computational procedures. In the concluding section, a brief discussion of the problem of appropriate model selection is presented, followed by a number of suggestions of essentially model-free alternative methods of use in the treatment of complex structured data such as occurs in genetic epidemiology.",
author = "S. Karlin and Cameron, {E. C.} and Ranajit Chakraborty",
year = "1983",
month = "1",
day = "1",
language = "English",
volume = "35",
pages = "695--732",
journal = "American Journal of Human Genetics",
issn = "0002-9297",
publisher = "Cell Press",
number = "4",

}

Karlin, S, Cameron, EC & Chakraborty, R 1983, 'Path analysis in genetic epidemiology: A critique', American Journal of Human Genetics, vol. 35, no. 4, pp. 695-732.

Path analysis in genetic epidemiology : A critique. / Karlin, S.; Cameron, E. C.; Chakraborty, Ranajit.

In: American Journal of Human Genetics, Vol. 35, No. 4, 01.01.1983, p. 695-732.

Research output: Contribution to journalReview article

TY - JOUR

T1 - Path analysis in genetic epidemiology

T2 - A critique

AU - Karlin, S.

AU - Cameron, E. C.

AU - Chakraborty, Ranajit

PY - 1983/1/1

Y1 - 1983/1/1

N2 - Path analysis, a form of general linear structural equation models, is used in studies of human genetics data to discern genetic, environmental, and cultural factors contributing to familial resemblance. It postulates a set of linear and additive parametric relationships between phenotypes and genetic and cultural variables and then essentially uses the assumption of multivariate normality to estimate and perform tests of hypothesis on parameters. Such an approach has been advocated for the analysis of genetic epidemiological data by D.C. Rao, N. Morton, C.R. Cloninger, L.J. Eaves, and W.E. Nance, among others. This paper reviews and evaluates the formulations, assumptions, methodological procedures, interpretations, and applications of path analysis. To give perspective, we begin with a discussion of path analysis as it occurs in the form of general linear causal models in several disciplines of the social sciences. Several specific path analysis models applied to lipoprotein concentrations, IQ, and twin data are then reviewed to keep the presentation self-contained. The bulk of the critical discussion that follows is directed toward the following four facets of path analysis: (1) coherence of model specification and applicability to data; (2) plausibility of modeling assumptions; (3) interpretability and utility of the model; and (4) validity of statistical and computational procedures. In the concluding section, a brief discussion of the problem of appropriate model selection is presented, followed by a number of suggestions of essentially model-free alternative methods of use in the treatment of complex structured data such as occurs in genetic epidemiology.

AB - Path analysis, a form of general linear structural equation models, is used in studies of human genetics data to discern genetic, environmental, and cultural factors contributing to familial resemblance. It postulates a set of linear and additive parametric relationships between phenotypes and genetic and cultural variables and then essentially uses the assumption of multivariate normality to estimate and perform tests of hypothesis on parameters. Such an approach has been advocated for the analysis of genetic epidemiological data by D.C. Rao, N. Morton, C.R. Cloninger, L.J. Eaves, and W.E. Nance, among others. This paper reviews and evaluates the formulations, assumptions, methodological procedures, interpretations, and applications of path analysis. To give perspective, we begin with a discussion of path analysis as it occurs in the form of general linear causal models in several disciplines of the social sciences. Several specific path analysis models applied to lipoprotein concentrations, IQ, and twin data are then reviewed to keep the presentation self-contained. The bulk of the critical discussion that follows is directed toward the following four facets of path analysis: (1) coherence of model specification and applicability to data; (2) plausibility of modeling assumptions; (3) interpretability and utility of the model; and (4) validity of statistical and computational procedures. In the concluding section, a brief discussion of the problem of appropriate model selection is presented, followed by a number of suggestions of essentially model-free alternative methods of use in the treatment of complex structured data such as occurs in genetic epidemiology.

UR - http://www.scopus.com/inward/record.url?scp=0020599524&partnerID=8YFLogxK

M3 - Review article

C2 - 6349335

AN - SCOPUS:0020599524

VL - 35

SP - 695

EP - 732

JO - American Journal of Human Genetics

JF - American Journal of Human Genetics

SN - 0002-9297

IS - 4

ER -

Karlin S, Cameron EC, Chakraborty R. Path analysis in genetic epidemiology: A critique. American Journal of Human Genetics. 1983 Jan 1;35(4):695-732.