Abstract
In analyzing causal claims, the most common evidentiary strategy is to use an experimental or quasi-experimental framework; holding all else constant, a treatment is varied and its effect on the outcome is determined. However, a second, quite distinct strategy is gaining prominence within the social sciences. Rather than mimic an experiment, researchers can identify causal relations by finding evidence for mechanisms that link cause and effect. In this chapter, we use Directed Acyclic Graphs (DAGs) to illustrate the power of using mechanisms. We show how mechanisms can aid in causal analysis by bringing additional variation to bear in instances where causal effects would otherwise not be identified. Specifically, we examine five generic situations where a focus on mechanisms using DAGs allows an analyst to warrant causal claims.
Authors listed in alphabetical order.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
- 3.
In situations where the effect of T and M interact in affecting Y and T affects Y independently of M, there are multiple possible definitions of what constitutes T’s direct and indirect effect since they will depend on the level of Y. We do not discuss this issue here further. Robins and Greenland (1992) provide an overview.
- 4.
Some philosophers of science, for example, believe the counterfactual criterion to be neither sufficient nor necessary for causation. We do not review these limitations here.
- 5.
- 6.
The literature on complex systems would describe a situation where (Fig. 14.1a) and (Fig. 14.1c) held as one of “supervenience”—that is, a case where a particular higher-order phenomena, here A → B, is the result of any one of two or more different separate lower-level processes, here A → M 1 → B or A → M 2 → B, and that there can be no change in the higher-order process without a change in the lower-order processes (Sawyer 2005).
- 7.
Elwert and Winship (2012) provide a lengthy discussion of these phenomena showing that a wide range of methodological biases are due to it. Furthermore, they describe a host of social science examples where this bias is present.
- 8.
Elwert and Winship (2012) show that in general the direction of the association may be either positive or negative when a collider has been conditioned on.
- 9.
For a formal definition of identifiability, see Pearl (2009, p. 77).
- 10.
A “descendant” of a variable is a variable that is either directly or indirectly caused by a variable.
- 11.
In a seminal paper, Angrist et al. (1996) show that when the effect of T on Y is heterogeneous, the traditional IV estimator only estimates the average causal effect of T on Y for those individuals whose behavior is shifted by Z, giving what they term the “local average treatment effect.” Of course in this situation, we would need to add to the DAG in Fig. 14.5 a variable W * that captured this heterogeneity and it is affecting both T and Y. If Z and W * were associated, then traditional IV identification would no longer hold as there would be an unblocked path going from Z to W * to Y. See section 7.5.1 of Morgan and Winship (2010) for further explanation.
- 12.
Criteria (1) and (2) need also to prohibit the possibility that by conditioning on a variable in order to block a path, one has not conditioned on a collider and thus induced a new unblocked path. See Elwert and Winship (2012) for a discussion of situations where if one is considering conditioning on a variable, one is “damned if you do and damned if you don’t.”
- 13.
As noted above, a causal DAG includes all common determinants of variables in the DAG.
- 14.
If Wages = f(Effort, Bias *) and Effort = g(Kids), then the convolution is found by substitution giving Wages = f(g(Kids), Bias *).
- 15.
- 16.
Arguably, there should also be a direct arrow in this diagram going from Kids to Effort, representing the fact that even with childcare, the presence of children might reduce a woman’s work effort. In order to keep things simple, however, we will assume here that childcare eliminates any tendency for woman with kids to work less. This might be a reasonable assumption if the organization itself provided childcare.
References
Anderson, D. J., Binder, M., & Krause, K. (2003). Motherhood wage penalty revisited: Experience, heterogeneity, work effort, and work-schedule flexibility. Industrial and Labor Relations Review, 56(2), 273.
Angrist, J. D., Imbens, G. W., & Rubin, D. B. (1996). Identification of causal effects using instrumental variables. Journal of the American Statistical Association, 91, 445–455.
Avin, C., Shpitser, I., & Pearl, J. (2005). Identifiability of path-specific effects. In Proceedings of International Joint Conference on Artificial Intelligence (pp. 357–363). Edinburgh, Scotland.
Bechtel, W. (2006). Discovering cell mechanisms. The creation of modern cell biology. New York: Cambridge University Press.
Bechtel, W. (2008). Mental mechanism. Philosophical perspectives on cognitive neuroscience. London: Routledge.
Bechtel, W., & Abrahamsen, A. (2005). Explanation: A mechanist alternative. Studies in History and Philosophy of Biological and Biomedical Sciences, 36, 421–441.
Becker, G. S. (1985). Human capital, effort, and the sexual division of labor. Journal of Labor Economics, 3(1), 33–58.
Bennett, A. (2008). Process tracing: A Bayesian perspective. In J. Box-Steffensmeier, H. E. Brady, & D. Collier (Eds.), The Oxford handbook of political methodology (pp. 217–270). Oxford: Oxford University Press.
Bollen, K. (2012). Instrumental variables in sociology and the social sciences. Annual Review of Sociology, 38, 37–72.
Budig, M. J., & England, P. (2001). The wage penalty for motherhood. American Sociological Review, 66, 204–225.
Bunge, M. (1997). Mechanism and explanation. Philosophy of the Social Sciences, 27, 410–465.
Bunge, M. (2004). How does it work? The search for explanatory mechanisms. Philosophy of the Social Sciences, 34, 182–210.
Collier, D., Brady, H. E., & Seawright, J. (2010). Sources of leverage in causal inference: Toward an alternative view of methodology. In H. E. Brady & D. Collier (Eds.), Rethinking social inquiry: Diverse tools, shared standards (2nd ed., pp. 161–199). Lanham: Rowman & Littlefield.
Correll, S. J., Benard, S., & Paik, I. (2007). Getting a job: Is there a motherhood penalty? American Journal of Sociology, 112(5), 1297–1339.
Craver, C. (2007). Explaining the brain. Mechanisms and the mosaic unity of neuroscience. Oxford: Oxford University Press.
Darden, L. (2002). Strategies for discovering mechanisms: Schema instantiation, modular subassembly, forward/backward chaining. Philosophy of Science, 69(S3), 354–365.
Darden, L. (2006). Reasoning in biological discoveries. Essays on mechanisms, interfield relations, and anomaly resolution. Cambridge, UK: Cambridge University Press.
Duncan, O. D., Featherman, D. L., & Duncan, B. (1972). Socioeconomic background and achievement. New York: Seminar.
Elster, J. (1989). Nuts and bolts for the social sciences. Cambridge, UK: Cambridge University Press.
Elster, J. (1999). Alchemies of the mind: Rationality and the emotions. Cambridge, UK: Cambridge University Press.
Elster, J. (2007). Explaining social behavior: More nuts and bolts for the social sciences. Cambridge: Cambridge University Press.
Elwert, F., & Winship, C. (2012). Endogenous selection. Unpublished paper, University of Wisconsin.
George, A. L., & Bennett, A. (2005). Case studies and theory development in the social sciences. Cambridge: MIT Press.
Gerring, J. (2007). Review article: The mechanistic worldview. Thinking inside the box. British Journal of Political Science, 38, 161–179.
Gerring, J. (2010). Causal mechanisms: Yes, but…. Comparative Political Studies, 43(11), 1499–1526.
Glennan, S. (2002). Rethinking mechanistic explanation. Philosophy of Science, 69, S342–S353.
Gross, N. (2009). A pragmatist theory of social mechanisms. American Sociological Review, 74(3), 358–379.
Hedström, P. (2005). Dissecting the social. On the principles of analytical sociology. Cambridge, UK: Cambridge University Press.
Hedstrom, P., & Bearman, P. (2009). The Oxford handbook of analytic sociology. Oxford: Oxford University Press.
Hedstrom, P., & Swedberg, R. (1996). Social mechanisms. Acta Sociologica, 39(3), 281–308.
Hedstrom, P., & Ylikoski, P. (2010). Causal mechanisms in the social sciences. Annual Review of Sociology, 36, 49–67.
Hume, D. (1739). A treatise on human nature. Oxford: Oxford University Press.
Imai, K., Keele, L., Tingley, D., & Yamamoto, T. (2011). Unpacking the black box of causality: Learning about causal mechanisms from experimental and observational studies. American Political Science Review, 105(4), 265–289.
Kendall, P. L., & Lazarsfeld, P. (1950). Problems of survey analysis. In R. K. Merton & P. F. Lazarsfeld (Eds.), Continuities in social research: Studies in the scope and methods of “The American Soldier” (pp. 133–196). Glencoe: Free Press.
King, G., Keohane, R. O., & Verba, S. (1994). Designing social inquiry: Scientific inference in qualitative research. Princeton: Princeton University Press.
Little, D. (1986). The scientific Marx. Minneapolis: University of Minnesota Press.
Little, D. (1991). Varieties of social explanation: An introduction to the philosophy of social science. Boulder: Westview.
Machamer, P., Darden, L., & Craver, C. F. (2000). Thinking about mechanisms. Philosophy of Science, 67(1), 1–25.
Mahoney, J. (2012). The logic of process tracing tests in the social sciences. Sociological Methods & Research, 41(4), 570–597.
Morgan, S. L., & Winship, C. (2007). Counterfactuals and causal inference: Methods and principles for social research. New York: Cambridge University Press.
Morgan, S. L., & Winship, C. (2010). Bringing context and variability back in to causal analysis. In H. Kincaid (Ed.), Oxford handbook of the philosophy of the social sciences (pp. 319–354). Oxford: Oxford University Press.
Pearl, J. (2000). Causality: Models, reasoning and inference. New York: Cambridge University Press.
Pearl, J. (2009). Causality: Models, reasoning, and inference (2nd ed.). New York: Cambridge University Press.
Pearl, J. (2010). On measurement bias in causal inference. In P. Grunwald, & P. Spirtes (Eds.), Proceedings of UAI (pp. 425–432). Corvallis: AUAI.
Pearl, J. (2012). The causal mediation formula—A guide to the assessment of pathways and mechanisms. Prevention Science, 13(4), 426–436.
Reskin, B. (2003). Including mechanisms in our models of ascriptive inequality. American Sociological Review, 68, 1–21.
Robins, J., & Greenland, S. (1992). Identifiability and exchangeability for direct and indirect effects. Epidemiology, 3(2), 143–155.
Salmon, W. C. (1984). Scientific explanation and causal structure of the world. Princeton: Princeton University Press.
Sawyer, R. K. (2005). Social emergence: Societies as complex systems. Cambridge: Cambridge University Press.
Sobel, M. E. (2008). Identification of causal parameters in randomized studies with mediating variables. Journal of Educational and Behavioral Statistics, 33(2), 230–251.
Swedberg, R. (2000). Max Weber and the idea of economic sociology. Princeton: Princeton University Press.
VanderWeele, T. (2009). Mediation and mechanism. European Journal of Epidemiology, 24, 217–224.
VanderWeele, T. J., & Robins, J. M. (2007). Directed acyclic graphs, sufficient causes, and the properties of conditioning on a common effect. American Journal of Epidemiology, 166(9), 1096–1104.
Winship, C., & Harding, D. (2008). A mechanism-based approach to the identification of age-period-cohort models. Sociological Methods & Research, 36(3), 363–340.
Woodward, J. (2002). What is a mechanism? A counterfactual account. Philosophy of Science, 69(3S), 366–378.
Woodward, J. (2005). Making things happen: A theory of causal explanation. New York: Oxford University Press.
Ylikoski, P. (2011). Social mechanisms and explanatory relevance. In P. Demeulenaere (Ed.), Analytical sociology and social mechanisms. New York: Cambridge University Press.
Acknowledgements
The authors wish to thank Steve Morgan, Judea Pearl, Tyler VanderWeele, Chris Muller, and Joe Krupnick for helpful comments. Of course, any remaining errors are the authors’ responsibility.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Knight, C.R., Winship, C. (2013). The Causal Implications of Mechanistic Thinking: Identification Using Directed Acyclic Graphs (DAGs). In: Morgan, S. (eds) Handbook of Causal Analysis for Social Research. Handbooks of Sociology and Social Research. Springer, Dordrecht. https://doi.org/10.1007/978-94-007-6094-3_14
Download citation
DOI: https://doi.org/10.1007/978-94-007-6094-3_14
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-007-6093-6
Online ISBN: 978-94-007-6094-3
eBook Packages: Humanities, Social Sciences and LawSocial Sciences (R0)