Handbook of Causal Analysis for Social Research pp 301-328 | Cite as

# Eight Myths About Causality and Structural Equation Models

## Abstract

Causality was at the center of the early history of structural equation models (SEMs) which continue to serve as the most popular approach to causal analysis in the social sciences. Through decades of development, critics and defenses of the capability of SEMs to support causal inference have accumulated. A variety of misunderstandings and myths about the nature of SEMs and their role in causal analysis have emerged, and their repetition has led some to believe they are true. Our chapter is organized by presenting eight myths about causality and SEMs in the hope that this will lead to a more accurate understanding. More specifically, the eight myths are the following: (1) SEMs aim to establish causal relations from associations alone, (2) SEMs and regression are essentially equivalent, (3) no causation without manipulation, (4) SEMs are not equipped to handle nonlinear causal relationships, (5) a potential outcome framework is more principled than SEMs, (6) SEMs are not applicable to experiments with randomized treatments, (7) mediation analysis in SEMs is inherently noncausal, and (8) SEMs do not test any major part of the theory against the data. We present the facts that dispel these myths, describe what SEMs can and cannot do, and briefly present our critique of current practice using SEMs. We conclude that the current capabilities of SEMs to formalize and implement causal inference tasks are indispensible; its potential to do more is even greater.

## Keywords

Latent Variable Structural Equation Model Causal Relation Causal Effect Path Analysis## Notes

### Acknowledgments

The authors would like to thank Shawn Bauldry, Stephen Cole, Keith Marcus, Cameron McIntosh, Stan Mulaik, Johannes Textor, and other researchers from SEMNET for their comments on and critiques of our chapter. Bollen’s work was partially supported by NSF SES 0617276. Pearl’s work was partially supported by grants from NIH #1R01 LM009961-01, NSF #IIS-0914211 and #IIS-1018922, and ONR #N000-14-09-1-0665 and #N00014-10-0933.

## References

- Ali, R., Richardson, T., & Spirtes, P. (2009). Markov equivalence for ancestral graphs.
*The Annals of Statistics, 37*, 2808–2837.CrossRefGoogle Scholar - Angrist, J. D., & Pischke, J.-S. (2009).
*Mostly harmless econometrics: An empiricist’s companion*. Princeton: Princeton University Press.Google Scholar - Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations.
*Journal of Personality and Social Psychology, 51*, 1173–1182.CrossRefGoogle Scholar - Baumrind, D. (1983). Specious causal attributions in the social sciences: The reformulated stepping-stone theory of heroin use as exemplar.
*Journal of Personality and Social Psychology, 45*, 1289–1298.CrossRefGoogle Scholar - Bentler, P. M., & Yuan, K.-H. (1999). Structural equation modeling with small samples: Test statistics.
*Multivariate Behavioral Research, 34*, 181–197.CrossRefGoogle Scholar - Berk, R. A. (2004).
*Regression analysis: A constructive critique*. Newbury Park: Sage.Google Scholar - Bhrolcháin, M. N., & Dyson, T. (2007). On causation in demography: Issues and illustrations.
*Population and Development Review, 33*, 1–36.CrossRefGoogle Scholar - Blalock, H. M., Jr. (1960). Correlation analysis and causal inferences.
*American Anthropologist, 62*, 624–631.CrossRefGoogle Scholar - Blalock, H. M., Jr. (1961).
*Causal inferences in nonexperimental research*. Chapel Hill: University of North Carolina Press.Google Scholar - Blalock, H. M., Jr. (1962). Four-variable causal models and partial correlations.
*American Journal of Sociology, 68*, 182–194.CrossRefGoogle Scholar - Blalock, H. M., Jr. (1963). Making causal inferences for unmeasured variables from correlations among indicators.
*American Journal of Sociology, 69*, 53–56.CrossRefGoogle Scholar - Blalock, H. M., Jr. (1969). Multiple indicators and the casual approach to measurement error.
*American Journal of Sociology, 75*, 264–272.CrossRefGoogle Scholar - Blalock, H. M., Jr. (1985).
*Causal models in the social sciences*. Hawthorne: Aldine de Gruyter.Google Scholar - Blau, P. M., & Duncan, O. D. (1967).
*The American occupational structure*. New York: The Free Press.Google Scholar - Bollen, K. A. (1987). Total, direct, and indirect effects in structural equation models.
*Sociological Methodology, 17*, 37–69.CrossRefGoogle Scholar - Bollen, K. A. (1989).
*Structural equations with latent variables*. New York: Wiley.Google Scholar - Bollen, K. A. (1990). Outlier screening and a distribution-free test for vanishing tetrads.
*Sociological Methods & Research, 19*, 80–92.CrossRefGoogle Scholar - Bollen, K. A. (1996). An alternative 2SLS estimator for latent variable models.
*Psychometrika, 61*, 109–121.CrossRefGoogle Scholar - Bollen, K. A. (2001). Two-stage least squares and latent variable models: Simultaneous estimation and robustness to misspecifications. In R. Cudeck, S. D. Toit, & D. Sörbom (Eds.),
*Structural equation modeling: Present and future, a Festschrift in honor of Karl Jöreskog*(pp. 119–138). Lincoln: Scientific Software.Google Scholar - Bollen, K. A. (2002). Latent variables in psychology and the social sciences.
*Annual Review of Psychology, 53*, 605–634.CrossRefGoogle Scholar - Bollen, K. A., & Bauer, D. J. (2004). Automating the selection of model-implied instrumental variables.
*Sociological Methods & Research, 32*, 425–452.CrossRefGoogle Scholar - Bollen, K. A., & Stine, R. (1990). Direct and indirect effects: Classical and bootstrap estimates of variability.
*Sociological Methodology, 20*, 115–140.CrossRefGoogle Scholar - Bollen, K. A., & Stine, R. A. (1993). Bootstrapping goodness-of-fit measures in structural equation modeling. In K. A. Bollen & J. S. Long (Eds.),
*Testing structural equation models*(pp. 111–135). Newbury Park: Sage.Google Scholar - Bollen, K. A., & Ting, K.-F. (1993). Confirmatory tetrad analysis.
*Sociological Methodology, 23*, 147–175.CrossRefGoogle Scholar - Brito, C., & Pearl, J. (2002). Generalized instrumental variables. In A. Darwiche & N. Friedman (Eds.),
*Proceedings of the eighteenth conference on uncertainty in artificial intelligence*(pp. 85–93). San Francisco: Morgan Kaufmann.Google Scholar - Cliff, N. (1983). Some cautions concerning the application of causal modeling methods.
*Multivariate Behavioral Research, 18*, 115–126.CrossRefGoogle Scholar - Costner, H. L. (1971). Utilizing causal models to discover flaws in experiments.
*Sociometry, 34*, 398–410.CrossRefGoogle Scholar - de Leeuw, J. (1985). Reviews.
*Psychometrika, 50*, 371–375.CrossRefGoogle Scholar - Duncan, O. D. (1966). Path analysis: Sociological examples.
*American Journal of Sociology, 72*, 1–16.CrossRefGoogle Scholar - Duncan, O. D. (1975).
*Introduction to structural equation models*. New York: Academic.Google Scholar - Everitt, B. S. (1984).
*An introduction to latent variable models*. New York: Chapman & Hall.CrossRefGoogle Scholar - Fisher, R. A. (1935).
*The design of experiments*. Edinburgh: Oliver and Boyd.Google Scholar - Folmer, H. (1981). Measurement of the effects of regional policy instruments by means of linear structural equation models and panel data.
*Environment and Planning A, 13*, 1435–1448.CrossRefGoogle Scholar - Fox, J. (1980). Effect analysis in structural equation models: Extensions and simplified methods of computation.
*Sociological Methods & Research, 9*, 3–28.CrossRefGoogle Scholar - Freedman, D. A. (1981). Pitfalls in large econometric models: A case study.
*Journal of Business, 54*, 479–500.CrossRefGoogle Scholar - Freedman, D. A. (1987). As others see us: A case study in path analysis.
*Journal of Educational and Behavioral Statistics, 12*, 101–128.CrossRefGoogle Scholar - Freedman, D. A. (2004). Graphical models for causation, and the identification problem.
*Evaluation Review, 28*, 267–293.CrossRefGoogle Scholar - Freedman, D. A. (2009). Diagnostics cannot have much power against general alternatives.
*International Journal of Forecasting, 25*, 833–839.CrossRefGoogle Scholar - Galles, D., & Pearl, J. (1998). An axiomatic characterization of causal counterfactuals.
*Foundations of Science*, 3(1), 151–182, Kluwer Academic Publishers, 1998.Google Scholar - Glymour, C. (1986). Statistics and causal inference: Comment: Statistics and metaphysics.
*Journal of the American Statistical Association, 81*, 964–966.Google Scholar - Glymour, C., Scheines, R., Spirtes, P., & Kelly, K. (1987).
*Discovering causal structure: Artificial intelligence, philosophy of science, and statistical modeling*. Orlando: Academic.Google Scholar - Goldberger, A. S. (1972). Structural equation methods in the social sciences.
*Econometrica, 40*, 979–1001.CrossRefGoogle Scholar - Goldberger, A. S. (1973). Structural equation models: An overview. In A. S. Goldberger & O. D. Duncan (Eds.),
*Structural equation models in the social sciences*(pp. 1–18). New York: Seminar Press.Google Scholar - Goldberger, A. S., & Duncan, O. D. (1973).
*Structural equation models in the social sciences*. New York: Seminar Press.Google Scholar - Goldthorpe, J. H. (2001). Causation, statistics, sociology.
*European Sociological Review, 17*, 1–20.CrossRefGoogle Scholar - Granger, C. W. (1969). Investigating causal relationships by econometric models and cross-spectral methods.
*Econometrica, 37*, 424–438.CrossRefGoogle Scholar - Greenland, S., Pearl, J., & Robins, J. M. (1999). Causal diagrams for epidemiologic research.
*Epidemiology, 10*, 37–48.CrossRefGoogle Scholar - Guttman, L. (1977). What is not what in statistics.
*Journal of the Royal Statistical Society: Series D (The Statistician), 26*, 81–107.Google Scholar - Haavelmo, T. (1943). The statistical implications of a system of simultaneous equations.
*Econometrica, 11*, 1–12.CrossRefGoogle Scholar - Halpern, J. (1998). Axiomatizing causal reasoning. In G. Cooper & S. Moral (Eds.),
*Uncertainty in artificial intelligence*(pp. 202–210). San Francisco: Morgan Kaufmann.Google Scholar - Hershberger, S. L. (2003). The growth of structural equation modeling: 1994–2001.
*Structural Equation Modeling, 10*, 35–46.CrossRefGoogle Scholar - Holland, P. W. (1986). Statistics and causal inference.
*Journal of the American Statistical Association, 81*, 945–970.CrossRefGoogle Scholar - Holland, P. W. (1995). Some reflections on Freedman’s critiques.
*Foundations of Science, 1*, 50–57.Google Scholar - Imai, K., Keele, L., & Yamamoto, T. (2010). Identification, inference, and sensitivity analysis for causal mediation effects.
*Statistical Science, 25*, 51–71.CrossRefGoogle Scholar - James, C. R., Mulaik, S. A., & Brett, J. M. (1982).
*Causal analysis: Assumptions, models, and data*. Beverly Hills: Sage.Google Scholar - Jöreskog, K. G. (1969). A general approach to confirmatory maximum likelihood factor analysis.
*Psychometrika, 34*, 183–202.CrossRefGoogle Scholar - Jöreskog, K. G. (1970). A general method for the analysis of covariance structures.
*Biometrika, 57*, 239–251.Google Scholar - Jöreskog, K. G. (1973). A general model for estimating a linear structural equation system. In A. S. Goldberger & O. D. Duncan (Eds.),
*Structural equation models in the social sciences*. New York: Seminar Press.Google Scholar - Jöreskog, K. G., & Sörbom, D. (1978).
*LISREL IV [Computer software]*. Chicago: Scientific Software International, Inc.Google Scholar - Jöreskog, K. G., & Sörbom, D. (1981).
*LISREL V [Computer software]*. Chicago: Scientific Software International, Inc.Google Scholar - Joffe, M. (2011). Principal stratification and attribution prohibition: Good ideas taken too far.
*International Journal of Biostatistics*,*7*, 1--13.Google Scholar - Joffe, M., Yang, W. P., & Feldman, H. I. (2010). Selective ignorability assumptions in causal inference.
*International Journal of Biostatistics*,*6*.Google Scholar - Kaufman, S., Kaufman, J. S., & Maclehose, R. F. (2009). Analytic bounds on causal risk differences in directed acyclic graphs involving three observed binary variables.
*Journal of Statistical Planning and Inference, 139*(10), 3473–3487.CrossRefGoogle Scholar - Kenny, D. A. (1979).
*Correlation and causality*. New York: Wiley.Google Scholar - Kirby, J. B., & Bollen, K. A. (2009). Using instrumental variable tests to evaluate model specification in latent variable structural equation models.
*Sociological Methodology, 39*, 327–355.CrossRefGoogle Scholar - Kyono, T. (2010).
*Commentator: A front-end user-interface module for graphical and structural equation modeling*(Tech. Rep. (R-364)). Los Angeles: Department of Computer Science, University of California. Available at http://ftp.cs.ucla.edu/pub/stat_ser/r364.pdf - Lee, J. L. (2012). Correlation and causation in the study of personality.
*European Journal of Personality, 26*, 372–390.CrossRefGoogle Scholar - Lindquist, M. A., & Sobel, M. E. (2011). Graphical models, potential outcomes and causal inference: Comment on Ramsey, Spirtes and Glymour.
*NeuroImage, 57*, 334–336.CrossRefGoogle Scholar - Long, J. S. (1983a).
*Confirmatory factor analysis*. Newbury Park: Sage.Google Scholar - Long, J. S. (1983b).
*Covariance structure models: An introduction to LISREL*. Newbury Park: Sage.Google Scholar - Miller, A. D. (1971). Logic of causal analysis: From experimental to nonexperimental designs. In H. M. Blalock Jr. (Ed.),
*Causal models in the social sciences*(pp. 273–294). Chicago: Aldine Atherton.Google Scholar - Morgan, S. L., & Winship, C. (2007).
*Counterfactuals and causal inference: Methods and principles for social research*. New York: Cambridge University Press.CrossRefGoogle Scholar - Mulaik, S. A. (2009).
*Linear causal modeling with structural equations*. Boca Raton: Chapman & Hall/CRC.CrossRefGoogle Scholar - Muthén, B. (1984). A general structural equation model with dichotomous, ordered categorical, and continuous latent variable indicators.
*Psychometrika, 49*, 115–132.CrossRefGoogle Scholar - Muthén, B. (2011).
*Applications of causally defined direct and indirect effects in mediation analysis using SEM in Mplus*(Tech. Rep.). Los Angeles: Graduate School of Education and Information Studies, University of California.Google Scholar - Pearl, J. (2000).
*Causality: Models, reasoning, and inference*(2nd ed., 2009). Cambridge: Cambridge University Press.Google Scholar - Pearl, J. (2001). Direct and indirect effects. In J. Breese & D. Koller (Eds.),
*Proceedings of the seventeenth conference on Uncertainty in Artificial Intelligence*(pp. 411–420). San Francisco: Morgan Kaufmann. http://ftp.cs.ucla.edu/pub/stat_ser/R273-U.pdf - Pearl, J. (2004, July). Robustness of causal claims.
*Proceedings of the 20th Conference on Uncertainty in Artificial Intelligence*(pp. 446–453). Banff, Canada.Google Scholar - Pearl, J. (2009a).
*Myth, confusion, and science in causal analysis*(Tech. Rep. (R-348)). UCLA Cognitive Systems Laboratory. http://ftp.cs.ucla.edu/pub/stat_ser/r348-warning.pdf - Pearl, J. (2009b). Causal inference in statistics: An overview.
*Statistics Surveys, 3*, 96–146.CrossRefGoogle Scholar - Pearl, J. (2010). The foundations of causal inference.
*Sociological Methodology, 40*, 75–149.CrossRefGoogle Scholar - Pearl, J. (2011a). Principal stratification – A goal or a tool?
*International Journal of Biostatistics, 7*(1), 20.Google Scholar - Pearl, J. (2011b).
*The causal mediation formula – A guide to the assessment of pathways and mechanisms*. http://ftp.cs.ucla.edu/pub/statser/r379.pdf, online.*Intervention Science*, doi: 10.1007/S11121-0270-1. - Pearl, J. (2011c). Forthcoming, Econometric Theory. http://ftp.cs.ucla.edu/pub/stat_ser/r391.pdf
- Pearl, J. (2012a). The causal foundation of structural equation modeling. In R. Hoyle (Ed.),
*Handbook of structural equation modeling*(pp. 68–91). Newbury Park: Sage.Google Scholar - Pearl, J. (2012b).
*Interpretable conditions for identifying direct and indirect effects*(Tech. Rep. (R-389)). http://ftp.cs.ucla.edu/pub/stat_ser/r389-tr.pdf - Petersen, M. (2011). Compound treatments, transportability, and the structural causal model: The power and simplicity of causal graphs.
*Epidemiology, 22*, 378–381.CrossRefGoogle Scholar - Provine, W. B. (1986).
*Sewall Wright and evolutionary biology*. Chicago: University of Chicago Press.Google Scholar - Robins, J. M. (1986). A new approach to causal inference in mortality studies with a sustained exposure period – Applications to control of the healthy workers survivor effect.
*Mathematical Modeling, 7*, 1393–1512.CrossRefGoogle Scholar - Robins, J. M. (2003). Semantics of causal (DAG) models and the identification of direct and indirect effects. In P. Green, N. Hjort, & S. Richardson (Eds.),
*Highly structured stochastic systems*(pp. 70–81). Oxford: Oxford University Press.Google Scholar - Robins, J., & Greenland, S. (1992). Identifiability and exchangeability for direct and indirect effects.
*Epidemiology, 3*, 143–155.CrossRefGoogle Scholar - Rubin, D. B. (1974). Estimating causal effects of treatments in randomized and nonrandomized studies.
*Journal of Educational Psychology, 66*, 688–701.CrossRefGoogle Scholar - Rubin, D. B. (2004). Direct and indirect causal effects via potential outcomes.
*Scandinavian Journal of Statistics, 31*, 162–170.Google Scholar - Rubin, D. B. (2009). Should observational studies be designed to allow lack of balance in covariate distributions across treatment groups.
*Statistics in Medicine, 28*, 1420–1423.CrossRefGoogle Scholar - Rubin, D. B. (2010). Reflections stimulated by the comments of Shadish (2010) and West and Thoemmes (2010).
*Psychological Methods, 15*, 38–46.CrossRefGoogle Scholar - Sargan, J. D. (1958). The estimation of economic relationships using instrumental variables.
*Econometrica, 26*, 393–415.CrossRefGoogle Scholar - Saris, W., & Stronkhorst, H. (1984).
*Causal modeling in nonexperimental research*. Amsterdam: Sociometric Research Foundation.Google Scholar - Satorra, A., & Bentler, P. M. (1994). Corrections to test statistics and standard errors in covariance structure analysis. In A. V. Eye & C. C. Clogg (Eds.),
*Latent variable analysis: Applications for developmental research*. Thousand Oaks: Sage.Google Scholar - Schnoll, R. A., Fang, C. Y., & Manne, S. L. (2004). The application of SEM to behavioral research in oncology: Past accomplishments and future opportunities.
*Structural Equation Modeling, 11*, 583–614.CrossRefGoogle Scholar - Schumacker, R. E., & Marcoulides, G. A. (Eds.). (1998).
*Interaction and nonlinear effects in structural equation modeling*. Mahway: Erlbaum.Google Scholar - Shadish, W. R., & Sullivan, K. J. (2012). Theories of causation in psychological science. In H. M. Cooper, P. M. Camic, D. L. Long, A. T. Panter, D. Rindskopf, & K. J. Sher (Eds.),
*APA handbook of research methods in psychology: Vol. 1. Foundations, planning, measures, and psychometrics*(pp. 3–17). Washington, DC: American Psychological Association.Google Scholar - Shah, R., & Goldstein, S. M. (2006). Use of structural equation modeling in operations management research: Looking back and forward.
*Journal of Operations Management, 24*, 148–169.CrossRefGoogle Scholar - Shapiro, A. (1986). Asymptotic theory of overparameterized structural models.
*Journal of the American Statistical Association, 81*, 142–149.CrossRefGoogle Scholar - Shpitser, I., & Pearl, J. (2008a). Complete identification methods for the causal hierarchy.
*Journal of Machine Learning, 9*, 1941–1979.Google Scholar - Shpitser, I., & Pearl, J. (2008b). Dormant independence. In
*Proceedings of the twenty-third conference on Artificial Intelligence*(pp. 1081–1087). Menlo Park: AAAI Press.Google Scholar - Shpitser, I., & Pearl, J. (2009). Effects of treatment on the treated: Identification and generalization. In J. Bilmes & A. Ng (Eds.),
*Proceedings of the twenty-fifth conference on uncertainty in artificial intelligence*. Montreal: AUAI Press.Google Scholar - Simon, H. A. (1954). Spurious correlation: A causal interpretation.
*Journal of the American Statistical Association, 49*, 467–479.Google Scholar - Sjolander, A. (2011). Reaction to Pearl’s critique of principal stratification.
*International Journal of Biostatistics, 7*, 1–5.CrossRefGoogle Scholar - Skrondal, A., & Rabe-Hesketh, S. (2005).
*Generalized latent variable modeling: Multilevel, longitudinal, and structural equation models*. Boca Raton: Chapman & Hall/CRC.Google Scholar - Sobel, M. E. (1986). Some new results on indirect effects and their standard errors in covariance structure models.
*Sociological Methodology, 16*, 159–186.CrossRefGoogle Scholar - Sobel, M. E. (2008). Identification of causal parameters in randomized studies with mediating variables.
*Journal of Educational and Behavioral Statistics, 33*, 230–251.CrossRefGoogle Scholar - Spirtes, P., Glymour, C., & Scheines, R. (2000).
*Causation, prediction, and search*(2nd ed.). Cambridge, MA: MIT Press.Google Scholar - Stolzenberg, R. M. (1979). The measurement and decomposition of causal effects in nonlinear and nonadditive models.
*Sociological Methodology, 11*, 459–488.CrossRefGoogle Scholar - VanderWeele, T. J. (2011). Principal stratification – Uses and limitations.
*International Journal of Biostatistics, 7*, 1–14.CrossRefGoogle Scholar - VanderWeele, T., & Vansteelandt, S. (2009). Conceptual issues concerning mediation, interventions and composition.
*Statistics and Its Interface, 2*, 457–468.Google Scholar - Verma, T., & Pearl, J. (1990). Equivalence and synthesis of causal models. In
*Uncertainty in artificial intelligence, Proceedings of the sixth conference*, Cambridge, MA.Google Scholar - White, H., & Chalak, K. (2009). Settable systems: An extension of Pearl’s causal model with optimization, equilibrium and learning.
*Journal of Machine Learning Research, 10*, 1759–1799.Google Scholar - Winship, C., & Mare, R. D. (1983). Structural equations and path analysis with discrete data.
*American Journal of Sociology, 89*, 54–110.CrossRefGoogle Scholar - Wright, S. S. (1918). On the nature of size factors.
*Genetics, 3*, 367–374.Google Scholar - Wright, S. S. (1921). Correlation and causation.
*Journal of Agricultural Research, 20*, 557–585.Google Scholar - Wright, P. G. (1928).
*The tariff on animal and vegetable oils*. New York: The MacMillan Company.Google Scholar - Wright, S. S. (1934). The method of path coefficients.
*Annals of Mathematical Statistics, 5*, 161–215.CrossRefGoogle Scholar