by Alan
The Affordable Care Act (ACA) of 2010 (aka "Obamacare") seeks to provide health-insurance coverage to an estimated 30 million Americans who were uninsured prior to the law's enactment. The two primary methods for doing so, both of which went into effect on October 1 of this year, are expanding eligibility for Medicaid, a long-existing government program for low-income individuals, and creating online marketplaces (or "exchanges") for people above the Medicaid income threshold to purchase private insurance (with federal subsidies available according to income level).
As I wrote on one of my other blogs back in August:
Starting next January [of 2014], individuals with income up to 133 percent of the poverty level will be eligible for Medicaid. The pre-ACA thresholds for Medicaid differ by state and by participant category (e.g., pregnant women, children, parents), but are sometimes as low as 50% of the poverty line.
As readers who have been following the Obamacare saga are no doubt aware, the U.S. Supreme Court ruled in 2012 that states could more readily opt out of the Medicaid expansion than had been intended by the ACA. Many states have indeed opted out, with governors' and legislatures' decisions heavily following party lines (here and here). Democrats have accepted the funding to expand their states' Medicaid eligibility and Republicans (largely) have not.
What struck me in reading about states' decisions on whether or not to accept the Medicaid expansion was the set of reasons cited by opponents for declining participation. Conservative arguments pertaining to an expanded federal-government role in health care and distaste in many states for increased spending on social programs (even though the federal government would largely cover the costs to states of expanding Medicaid) did not surprise me. What did surprise me was the claim that Medicaid is actually harmful to its participants and that people would be better off uninsured than with Medicaid.
Indeed, there are several published studies (many of which are compiled here) appearing to show Medicaid patients faring worse than their uninsured counterparts on health outcomes such as mortality, heart attacks, and timely diagnoses of serious illnesses. Perhaps the most widely publicized and discussed is a 2010 investigation (often referred to as the "University of Virginia study") by LaPar and colleagues entitled "Primary Payer Status Affects Mortality for Major Surgical Operations" (full-text).
In that article, which is based on a very large national data set containing statistics on surgical procedures such as hip replacement and coronary bypass, one finds that whereas uninsured patients had 74% greater odds of suffering in-hospital mortality than their privately insured counterparts, Medicaid patients had 97% greater odds of dying in the hospital than the private-insurance group (Table 6). This is, of course, a correlational or observational research design, linking the type of insurance a patient happened to have with his or her quality of surgical recovery. And with such a research design, patients will differ in many other ways than just their type of insurance coverage. Short of randomly assigning people to type of insurance (more on that later), we are always left with some degree of ambiguity in determining cause and effect.
In the Virginia surgical study, for example, the uninsured appeared actually to be quite wealthy, with roughly 60% in the two highest income categories (based on median incomes within the ZIP codes in which people live): 31.1% in the $45,000-or-greater category and 27.8% within $35,000-44,999 (Table 2). In contrast, only a combined 31% of Medicaid patients lived in the two highest-income sets of ZIP codes, with Medicaid patients clustered mainly (41.3%) in the lowest quartile (less than $25,000). The statistical analysis controlled for patient income, but to the extent the income measure did not capture all facets of patients' personal socioeconomic statuses, income may not have been as potent a control variable as possible (the authors noted that education and nutrition were among variables not included).
Interestingly, Avik Roy, a prominent Medicaid critic, acknowledges that unmeasured aspects associated with income may have played a role in the outcome of the Virginia study:
Another key element to consider is that many of the uninsured are not poor... These individuals are wealthy and/or healthy enough that they have decided to forego insurance. Though the Virginia study corrects for income status and other social factors, the fact that these patients are more capable of paying directly for their own care, at the prevailing rate, means that physicians are more willing to see them.
Another reason to be cautious about making a causal conclusion from the Virginia study that Medicaid is actively harmful to its subscribers is the lack of a well-established mechanism leading from Medicaid to poor health. As another Obamacare critic, Glenn Harlan Reynolds, acknowledges in a USA Today column, "Why Medicaid recipients do worse isn't entirely clear..." (Reynolds does suggest a few possible mechanisms, such as, "Uninsured patients probably go straight to the Emergency Room or to a free clinic, while Medicaid recipients may waste precious days, weeks, or months trying to navigate the bureaucracy." This particular conjecture turns out not to be supported, as Medicaid patients are especially likely to use the ER.)
Avik Roy, in his article cited above, also suggests potential mechanisms:
...the answer almost certainly begins with access to care. Medicaid’s extreme underpayment of doctors and hospitals leads fewer and fewer health-care providers to offer their services to Medicaid beneficiaries.
However, the evidence Roy states for this proposition pertains to both Medicaid patients and the uninsured.
Into the debate charges Austin Frakt, part of an ensemble of bloggers at the The Incidental Economist, whose training (collectively) includes medicine, social sciences, and statistics. Frakt advocates the use of a statistical technique called instrumental variables (IV) to deal with the correlational nature of most Medicaid-related studies and the inevitable omission of potentially relevant control variables and, in fact, he wrote a whole series of postings on Medicaid and instrumental variables. In one posting, Frakt writes:
There are observational studies that purport... that Medicaid coverage is worse or no better than being uninsured. One cannot draw such conclusions from such studies if they do not control for the unobservable factors that drive Medicaid enrollment. Causal inference requires appropriate techniques. Even a regression with lots of controls, even propensity score analysis, is insufficient in this area of study.
In another posting, Frakt writes:
Avik [Roy] dismisses IV as a “fudge factor,” casually and erroneously discrediting a vast amount of mainstream work by economists and several entire sub-disciplines. Since IV is a generalization of the concepts that underlie randomized controlled trials (differing in degree, but not in spirit, from purposeful randomization), and can be used to rehabilitate a trial with contaminated groups — a not infrequent occurrence – it is unwise to trivialize IV and what it can do.
According to Will Shadish and Tom Cook, research methodologists who write mainly in the areas of psychology, sociology, and program evaluation, "An instrumental variable is related to the treatment but not to the outcome except through its relationship on the treatment" (2009, p. 613). As suggested in some of Frakt's postings, for example, variation in states' Medicaid eligibility thresholds presumably would affect Medicaid enrollment, but would affect health only through Medicaid subscription. In a particularly useful posting, Frakt walks readers through a study of insurance and HIV treatment by Goldman et al., which used instrumental variables. [Here's another good explanation of instrumental variables, by David Kenny, which I forgot to include in my original posting.]
Frakt summarizes his Medicaid-Instrumental Variable series with this conclusion:
...there is no credible evidence that Medicaid results in worse or equivalent health outcomes as being uninsured. That is Medicaid improves health. It certainly doesn’t improve health as much as private insurance, but the credible evidence to date–that using sound techniques that can control for the self-selection into the program–strongly suggests Medicaid is better for health than no insurance at all.
As hinted above, opportunities for random-assignment experiments of Medicaid effectiveness occasionally do exist. Reference to true experiments would appear to be a good check on the validity of instrumental-variable studies purporting to substitute for randomized studies. Because Oregon's Medicaid program had been over-subscribed, the state used a lottery system to determine which eligible individuals were allowed to enroll in Medicaid and which were not. The random-assignment element replicates a traditional experiment, with the groups who were and were not admitted into Medicaid available for comparison of their health status over the following years. However, even a study that, in principle, is random-assignment can suffer from deficiencies, such as the incomplete participation within conditions in Oregon.
Still, this past May, two-year follow-up results of the Oregon Medicaid experiment were reported. Though the health benefits of being on Medicaid (vs. no insurance) were modest, there were some differences. Being on Medicaid led to: reduced probability of catastrophic health expenses, greater diagnosis of diabetes, and better treatment for depression. Various perspectives on the Oregon study are available here and here, as well as by searching on The Incidental Economist for Oregon Medicaid (the bloggers there wrote a huge number of posts about the study). Also worth noting, briefly and in conclusion, are other quasi-experimental methods that have been used to study the effectiveness of Medicaid: difference-in-difference and regression-discontinuity design.
By now it should be clear that trying to infer causality as to whether Medicaid leaves its holders better off, worse off, or unchanged is complex business. Still, there appears to be some common ground between analysts who, for the most part, interpret the Medicaid studies differently. Ultimately, Roy concludes his above-cited article with the acknowledgement that:
There is, doubtless, a level of poverty at which Medcaid is better than nothing at all. But most people can afford to take on more responsibility for their own care, and indeed would be far better off doing so.
References
Shadish, W. R., & Cook, T. D. (2009). The renaissance of field experimentation in evaluating interventions. Annual Review of Psychology 60, 607-629.
Friday, November 15, 2013
Monday, August 13, 2012
Causal Mediation Conference in Belgium
The Center for Statistics at Belgium's Ghent University will present the symposium
"Causal Mediation Analysis" on January 28-29, 2013, at Het Pand, Gent, Belgium. Further information is available here. According to a notice on the event, "This meeting aims to bridge the gap between traditional mediation analysis building on the famous work of Baron and Kenny and state-of-the-art causal mediation analysis. The idea is to discuss recent developments between methodological researchers and to bring them to the wider research community in a non-technical way."
Sunday, March 25, 2012
Michael Nielsen Offers (Relatively) Accessible Explanation of Pearl's "Causal Calculus"
by Alan
Judea Pearl, a UCLA professor of computer science, is one of the world's leading thinkers -- if not the leading thinker -- on conceptual approaches to causal inference. He is author of the book Causality and of numerous articles and presentations. He also operates the UCLA Causality Blog, a link to which appears in the left-hand column of the present page. On top of all this, Pearl recently garnered the Association for Computing Machinery (ACM) Turing Award for his contributions to artificial intelligence.
"Accessible" is not a word I would use to describe Pearl's writings, however. I have previously described the level of Pearl's writing as "quite frankly, well over my head." Heavy with logic symbols, Pearl's texts would, I suspect, challenge even many well-educated students of causality.
Fortunately for those of us seeking greater understanding of Pearl's ideas, Michael Nielsen has written an article trying to explain Pearl's "causal calculus" to a wider audience. I couldn't understand everything Nielsen wrote, but in relative terms, I found his exposition easier to grasp than Pearl's.
Fairly early on, Nielsen introduces the familiar example of smoking and lung cancer to discuss what conclusions can be drawn from correlational (observation) vs. randomized-controlled research designs (he seems to use the word "experimental" generically for any empirical investigation, specifying with terms such as "intervention" or "randomized controlled" when he means that participants are randomly assigned to conditions). Noting that human participants cannot ethically be randomly assigned to smoke cigarettes, Nielsen tantalizes the reader as follows:
We’ll see that even without doing a randomized controlled experiment it’s possible (with the aid of some reasonable assumptions) to infer what the outcome of a randomized controlled experiment would have been, using only relatively easily accessible experimental data, data that doesn’t require experimental intervention to force people to smoke or not, but which can be obtained from purely observational studies.
The main points I gleaned from Nielsen's piece were that (a) we can learn more than I previously thought simply from diagramming hypothetical causal relations between variables as in structural equation modeling or path analysis; and (b) one's conceptual model can be translated into conditional probability statements (i.e., given x, what is the probability of y) that potentially can be manipulated to answer causal questions without a randomized experiment. As Nielsen explains:
...Pearl had what turns out to be a very clever idea: to imagine a hypothetical world in which it really is possible to force someone to (for example) smoke, or not smoke. In particular, he introduced a conditional causal probability p(cancer|do(smoking)), which is the conditional probability of cancer in this hypothetical world. This should be read as the (causal conditional) probability of cancer given that we “do” smoking, i.e., someone has been forced to smoke in a (hypothetical) randomized experiment.
Now, at first sight this appears a rather useless thing to do. But what makes it a clever imaginative leap is that although it may be impossible or impractical to do a controlled experiment to determine p(cancer|do(smoking)), Pearl was able to establish a set of rules – a causal calculus – that such causal conditional probabilities should obey. And, by making use of this causal calculus, it turns out to sometimes be possible to infer the value of probabilities such as p(cancer|do(smoking)), even when a controlled, randomized experiment is impossible.
Returning to the lung-cancer example, it is theoretically possible that smoking leads directly to lung cancer or that an unobserved third variable causes both smoking and lung cancer (also, lung cancer may cause people to begin smoking, but that seems implausible). As Nielsen discusses, we can insert a fourth variable, namely particulate lung residue ("tar"), between smoking and lung cancer in the proposed causal sequence. This inclusion helps us partially break the connection between the hidden third variable and the other variables. Argues Nielsen: "But if the hidden causal factor is genetic, as the tobacco companies argued was the case, then it seems highly unlikely that the genetic factor caused tar in the lungs, except by the indirect route of causing those people to smoke."
Through manipulations such as the above: "the causal calculus lets us do something that seems almost miraculous: we can figure out the probability that someone would get cancer given that they are in the smoking group in a randomized controlled experiment, without needing to do the randomized controlled experiment. And this is true even though there may be a hidden causal factor underlying both smoking and cancer."
Ultimately, the manipulation of equations can lead to a formula to estimate the conditional probability of developing cancer given random assignment to a smoking condition, p(cancer|do(smoking)), as a function of "quantities which may be observed directly from experimental data, and which don’t require intervention to do a randomized, controlled experiment" (see Equation 5 in Nielsen's article). For any given problem, such non-intervention-based probabilities to plug into the equation may or may not be available.
Nielsen concludes the article by exploring possible future directions in the study of causality. For those interested in causal inference without randomized-controlled studies, Nielsen's article is a must-read.
Judea Pearl, a UCLA professor of computer science, is one of the world's leading thinkers -- if not the leading thinker -- on conceptual approaches to causal inference. He is author of the book Causality and of numerous articles and presentations. He also operates the UCLA Causality Blog, a link to which appears in the left-hand column of the present page. On top of all this, Pearl recently garnered the Association for Computing Machinery (ACM) Turing Award for his contributions to artificial intelligence.
"Accessible" is not a word I would use to describe Pearl's writings, however. I have previously described the level of Pearl's writing as "quite frankly, well over my head." Heavy with logic symbols, Pearl's texts would, I suspect, challenge even many well-educated students of causality.
Fortunately for those of us seeking greater understanding of Pearl's ideas, Michael Nielsen has written an article trying to explain Pearl's "causal calculus" to a wider audience. I couldn't understand everything Nielsen wrote, but in relative terms, I found his exposition easier to grasp than Pearl's.
Fairly early on, Nielsen introduces the familiar example of smoking and lung cancer to discuss what conclusions can be drawn from correlational (observation) vs. randomized-controlled research designs (he seems to use the word "experimental" generically for any empirical investigation, specifying with terms such as "intervention" or "randomized controlled" when he means that participants are randomly assigned to conditions). Noting that human participants cannot ethically be randomly assigned to smoke cigarettes, Nielsen tantalizes the reader as follows:
We’ll see that even without doing a randomized controlled experiment it’s possible (with the aid of some reasonable assumptions) to infer what the outcome of a randomized controlled experiment would have been, using only relatively easily accessible experimental data, data that doesn’t require experimental intervention to force people to smoke or not, but which can be obtained from purely observational studies.
The main points I gleaned from Nielsen's piece were that (a) we can learn more than I previously thought simply from diagramming hypothetical causal relations between variables as in structural equation modeling or path analysis; and (b) one's conceptual model can be translated into conditional probability statements (i.e., given x, what is the probability of y) that potentially can be manipulated to answer causal questions without a randomized experiment. As Nielsen explains:
...Pearl had what turns out to be a very clever idea: to imagine a hypothetical world in which it really is possible to force someone to (for example) smoke, or not smoke. In particular, he introduced a conditional causal probability p(cancer|do(smoking)), which is the conditional probability of cancer in this hypothetical world. This should be read as the (causal conditional) probability of cancer given that we “do” smoking, i.e., someone has been forced to smoke in a (hypothetical) randomized experiment.
Now, at first sight this appears a rather useless thing to do. But what makes it a clever imaginative leap is that although it may be impossible or impractical to do a controlled experiment to determine p(cancer|do(smoking)), Pearl was able to establish a set of rules – a causal calculus – that such causal conditional probabilities should obey. And, by making use of this causal calculus, it turns out to sometimes be possible to infer the value of probabilities such as p(cancer|do(smoking)), even when a controlled, randomized experiment is impossible.
Returning to the lung-cancer example, it is theoretically possible that smoking leads directly to lung cancer or that an unobserved third variable causes both smoking and lung cancer (also, lung cancer may cause people to begin smoking, but that seems implausible). As Nielsen discusses, we can insert a fourth variable, namely particulate lung residue ("tar"), between smoking and lung cancer in the proposed causal sequence. This inclusion helps us partially break the connection between the hidden third variable and the other variables. Argues Nielsen: "But if the hidden causal factor is genetic, as the tobacco companies argued was the case, then it seems highly unlikely that the genetic factor caused tar in the lungs, except by the indirect route of causing those people to smoke."
Through manipulations such as the above: "the causal calculus lets us do something that seems almost miraculous: we can figure out the probability that someone would get cancer given that they are in the smoking group in a randomized controlled experiment, without needing to do the randomized controlled experiment. And this is true even though there may be a hidden causal factor underlying both smoking and cancer."
Ultimately, the manipulation of equations can lead to a formula to estimate the conditional probability of developing cancer given random assignment to a smoking condition, p(cancer|do(smoking)), as a function of "quantities which may be observed directly from experimental data, and which don’t require intervention to do a randomized, controlled experiment" (see Equation 5 in Nielsen's article). For any given problem, such non-intervention-based probabilities to plug into the equation may or may not be available.
Nielsen concludes the article by exploring possible future directions in the study of causality. For those interested in causal inference without randomized-controlled studies, Nielsen's article is a must-read.
Tuesday, August 30, 2011
Judea Pearl and colleagues have launched the Journal of Causal Inference, to be published by the Berkeley Electronic Press.
Sunday, August 7, 2011
Causal-Inference References from SEMNET
by Alan
Over at the Structural Equation Modeling discussion listserve (SEMNET), participants lately have recommended several recent articles and resources on causal inference (and related topics) with non-experimental (correlational) research designs. For benefit of the larger research community, I have listed these materials below.
I've looked over some of these articles and they seem to vary in the prior training assumed. Some would seem accessible for social scientists without elaborate mathematical training, whereas others refer extensively to more sophisticated math (e.g., matrix algebra). The Antonakis et al. piece, in particular, appears to provide a (mostly) non-technical overview.
Three specific topics are covered in many of the articles:
Omitted-variable bias (or specification error, more generally), which is all-important to causal inference, due to the "third-variable" issue.
Propensity-score modeling, which already has been discussed extensively on this blog (e.g., here, here, and here).
The use of instrumental variables.
I am particularly impressed by the range of academic discliplines from which these articles arise. Thanks to those who contributed these items to SEMNET!
----------------------------------------------------------------------------------
Antonakis, J., Bendahan, S., Jacquart, P., & Lalive, R. (2010). On making causal claims: A review and recommendations. The Leadership Quarterly, 21, 1086-1120.
Austin, P.C. (2011). A tutorial and case study in propensity score analysis: An application to estimating the effect of in-hospital smoking cessation counseling on mortality. Multivariate Behavioral Research, 46, 119-151.
Beguin, J., Pothier, D., & Côté, S.D. (2011). Deer browsing and soil disturbance induce cascading effects on plant communities: a multilevel path analysis. Ecological Applications, 21, 439–451.
Bollen, K.A., Kirby, J.B., Curran, P.J., Paxton, P.M., & Chen, F. (2007). Latent variable models under misspecification: Two-Stage Least Squares (2SLS) and Maximum Likelihood (ML) estimators. Sociological Methods & Research, 36, 48-86.
Bollen, K.A., & Bauer, D.J. (2004). Automating the selection of model-implied instrumental variables. Sociological Methods & Research, 32, 425-452.
Bollen, K.A., & Maydeu-Olivares, A. (2007). A polychoric instrumental variable (PIV) estimator for structural equation models with categorical variables. Psychometrika, 72, 309-326.
Clarke, K. (2005). The Phantom Menace: Omitted variable bias in econometric research. Conflict Management and Peace Science, 22, 341-352.
Clarke, K. (2009). Return of the Phantom Menace: Omitted variable bias in econometric research. Conflict Management and Peace Science, 26, 46-66.
Coffman, D.L. (2011). Estimating causal effects in mediation analysis using propensity scores. Structural Equation Modeling, 18, 357-369.
Freedman, D.A., Collier, D., Sekhon, J.S., & Stark, P.B. (Eds.). (2009). Statistical models and causal inference: A dialogue with the social sciences. Cambridge University Press. ISBN: 978-0521123909.
Frosch, C.A., & Johnson-Laird, P.N. (2011). Is everyday causation deterministic or probabilistic? Acta Psychologica, 137, 280-291.
Hancock, G. R., & Harring, J. R. (2011, May). Using phantom variables in structural equation modeling to assess model sensitivity to external misspecification. Paper presented at the Modern Modeling Methods conference, Storrs, CT. (Hancock webpage to request copy.)
Hoshino, T. (2008). A Bayesian propensity score adjustment for latent variable modeling and MCMC algorithm. Computational Statistics & Data Analysis, 52, 1413-1429.
Kirby, J.B., & Bollen, K.A. (2009). Using instrumental variable (IV) tests to evaluate model specification in latent variable structural equation models. Sociological Methodology, 39, 327–355. (Public copy)
Mahoney, J. (2008). Toward a unified theory of causality. Comparative Political Studies, 41, 412-436.
Markus, K.A. (2011). Mulaik on atomism, contraposition and causation. Quality and Quantity. Online First (subscription needed), http://www.springerlink.com/content/r754405614228w0v/
Markus, K.A. (2011). Real causes and ideal manipulations: Pearl's theory of causal inference from the point of view of psychological resarch methods. In P. McKay Illari, F. Russo & J. Williamson (Eds.),Causality in the sciences (pp. 240-269). Oxford, UK: Oxford University Press. (Errata)
Pearl, J. (2010). On a class of bias-amplifying variables that endanger effect estimates. Technical Report R-356. In P. Grunwald & P. Spirtes (Eds.), Proceedings of UAI, 417-424. Corvallis, OR: AUAI.
Pearl, J. (2011, August). The causal foundations of structural equation modeling. UCLA Cognitive Systems Laboratory, Technical Report (R-370), http://ftp.cs.ucla.edu/pub/stat_ser/r370.pdf. Chapter for R. H. Hoyle (Ed.), Handbook of structural equation modeling. New York: Guilford Press.
Shadish, W.R., & Steiner, P.M. (2010). A primer on propensity score analysis. Newborn & Infant Nursing Review, 10, 19-26.
Shipley, B. (2009). Confirmatory path analysis in a generalized multilevel context. Ecology, 90, 363-368.
Shipley, B. The Causal Toolbox: A collection of programs for testing or exploring causal relationships [website]. http://pages.usherbrooke.ca/jshipley/recherche/book.htm
Spector, P.E., & Brannick, M.T. (2011). Methodological urban legends: The misuse of statistical control variables. Organizational Research Methods, 14, 287-305.
Steiner, P.M., Cook, T.D., Shadish, W.R., & Clark, M.H. (2010). The importance of covariate selection in controlling for selection bias in observational studies. Psychological Methods, 15, 250-267.
Thoemmes, F.J., & Kim, E.S. (2011). A systematic review of propensity score methods in the social sciences. Multivariate Behavioral Research, 46, 90-118.
Over at the Structural Equation Modeling discussion listserve (SEMNET), participants lately have recommended several recent articles and resources on causal inference (and related topics) with non-experimental (correlational) research designs. For benefit of the larger research community, I have listed these materials below.
I've looked over some of these articles and they seem to vary in the prior training assumed. Some would seem accessible for social scientists without elaborate mathematical training, whereas others refer extensively to more sophisticated math (e.g., matrix algebra). The Antonakis et al. piece, in particular, appears to provide a (mostly) non-technical overview.
Three specific topics are covered in many of the articles:
I am particularly impressed by the range of academic discliplines from which these articles arise. Thanks to those who contributed these items to SEMNET!
----------------------------------------------------------------------------------
Antonakis, J., Bendahan, S., Jacquart, P., & Lalive, R. (2010). On making causal claims: A review and recommendations. The Leadership Quarterly, 21, 1086-1120.
Austin, P.C. (2011). A tutorial and case study in propensity score analysis: An application to estimating the effect of in-hospital smoking cessation counseling on mortality. Multivariate Behavioral Research, 46, 119-151.
Beguin, J., Pothier, D., & Côté, S.D. (2011). Deer browsing and soil disturbance induce cascading effects on plant communities: a multilevel path analysis. Ecological Applications, 21, 439–451.
Bollen, K.A., Kirby, J.B., Curran, P.J., Paxton, P.M., & Chen, F. (2007). Latent variable models under misspecification: Two-Stage Least Squares (2SLS) and Maximum Likelihood (ML) estimators. Sociological Methods & Research, 36, 48-86.
Bollen, K.A., & Bauer, D.J. (2004). Automating the selection of model-implied instrumental variables. Sociological Methods & Research, 32, 425-452.
Bollen, K.A., & Maydeu-Olivares, A. (2007). A polychoric instrumental variable (PIV) estimator for structural equation models with categorical variables. Psychometrika, 72, 309-326.
Clarke, K. (2005). The Phantom Menace: Omitted variable bias in econometric research. Conflict Management and Peace Science, 22, 341-352.
Clarke, K. (2009). Return of the Phantom Menace: Omitted variable bias in econometric research. Conflict Management and Peace Science, 26, 46-66.
Coffman, D.L. (2011). Estimating causal effects in mediation analysis using propensity scores. Structural Equation Modeling, 18, 357-369.
Freedman, D.A., Collier, D., Sekhon, J.S., & Stark, P.B. (Eds.). (2009). Statistical models and causal inference: A dialogue with the social sciences. Cambridge University Press. ISBN: 978-0521123909.
Frosch, C.A., & Johnson-Laird, P.N. (2011). Is everyday causation deterministic or probabilistic? Acta Psychologica, 137, 280-291.
Hancock, G. R., & Harring, J. R. (2011, May). Using phantom variables in structural equation modeling to assess model sensitivity to external misspecification. Paper presented at the Modern Modeling Methods conference, Storrs, CT. (Hancock webpage to request copy.)
Hoshino, T. (2008). A Bayesian propensity score adjustment for latent variable modeling and MCMC algorithm. Computational Statistics & Data Analysis, 52, 1413-1429.
Kirby, J.B., & Bollen, K.A. (2009). Using instrumental variable (IV) tests to evaluate model specification in latent variable structural equation models. Sociological Methodology, 39, 327–355. (Public copy)
Mahoney, J. (2008). Toward a unified theory of causality. Comparative Political Studies, 41, 412-436.
Markus, K.A. (2011). Mulaik on atomism, contraposition and causation. Quality and Quantity. Online First (subscription needed), http://www.springerlink.com/content/r754405614228w0v/
Markus, K.A. (2011). Real causes and ideal manipulations: Pearl's theory of causal inference from the point of view of psychological resarch methods. In P. McKay Illari, F. Russo & J. Williamson (Eds.),Causality in the sciences (pp. 240-269). Oxford, UK: Oxford University Press. (Errata)
Pearl, J. (2010). On a class of bias-amplifying variables that endanger effect estimates. Technical Report R-356. In P. Grunwald & P. Spirtes (Eds.), Proceedings of UAI, 417-424. Corvallis, OR: AUAI.
Pearl, J. (2011, August). The causal foundations of structural equation modeling. UCLA Cognitive Systems Laboratory, Technical Report (R-370), http://ftp.cs.ucla.edu/pub/stat_ser/r370.pdf. Chapter for R. H. Hoyle (Ed.), Handbook of structural equation modeling. New York: Guilford Press.
Shadish, W.R., & Steiner, P.M. (2010). A primer on propensity score analysis. Newborn & Infant Nursing Review, 10, 19-26.
Shipley, B. (2009). Confirmatory path analysis in a generalized multilevel context. Ecology, 90, 363-368.
Shipley, B. The Causal Toolbox: A collection of programs for testing or exploring causal relationships [website]. http://pages.usherbrooke.ca/jshipley/recherche/book.htm
Spector, P.E., & Brannick, M.T. (2011). Methodological urban legends: The misuse of statistical control variables. Organizational Research Methods, 14, 287-305.
Steiner, P.M., Cook, T.D., Shadish, W.R., & Clark, M.H. (2010). The importance of covariate selection in controlling for selection bias in observational studies. Psychological Methods, 15, 250-267.
Thoemmes, F.J., & Kim, E.S. (2011). A systematic review of propensity score methods in the social sciences. Multivariate Behavioral Research, 46, 90-118.
Tuesday, February 1, 2011
Sunday, December 19, 2010
Correlation, Causality, and Parenting Studies
Alan and Bo are both quoted in this new article from Brain, Child magazine. In addition to discussing substantive issues of causal inference, author Katy Read also probes the chain of diffusion of social-science research.
The chain begins, of course, with the investigators who conducted the research. Those who conduct correlational studies typically include a statement of limitations at the end of their articles, noting that the findings are open to alternative causal interpretations. In less-guarded moments, however, even research scientists will use phraseology that implies a preferred causal direction.
Universities, research institutes, and/or professional organizations may then issue press releases on a particular study. Ultimately, a research finding may make it into the media. At each reporting step removed from the (methodologically trained) scientific investigators, therefore, statements of caution regarding causality are less likely to appear.
The chain begins, of course, with the investigators who conducted the research. Those who conduct correlational studies typically include a statement of limitations at the end of their articles, noting that the findings are open to alternative causal interpretations. In less-guarded moments, however, even research scientists will use phraseology that implies a preferred causal direction.
Universities, research institutes, and/or professional organizations may then issue press releases on a particular study. Ultimately, a research finding may make it into the media. At each reporting step removed from the (methodologically trained) scientific investigators, therefore, statements of caution regarding causality are less likely to appear.
Saturday, November 27, 2010
Studying Personality as a Causal Agent
by Alan
Angela Lee Duckworth, Eli Tsukayama, and Henry May have published an article entitled, "Establishing Causality Using Longitudinal Hierarchical Linear Modeling: An Illustration Predicting Achievement From Self-Control" (abstract) in the October 2010 issue of the new journal Social Psychological and Personality Science.
The authors are interested in the personality trait of self-control (which encompasses such abilities as persistence and delay of gratification) and whether it can actually be shown to cause academic achievement (GPA) during the years from fifth to eighth grade. Duckworth and colleagues acknowledge the difficulty early on:
"Manipulating personality in a random-assignment experiment could, in theory, establish its causal role for later outcomes but, alas, personality is not easily manipulated. To our knowledge, no empirical investigation to date has successfully manipulated trait-level self-control and measured subsequent effects on life outcomes" (p. 311).
The authors also carefully review the arguments for why longitudinal predictive models using path-analysis and structural equation modeling (controlling for prior levels of the dependent variable), though an improvement over cross-sectional correlations, are still vulnerable to unmeasured third variables. "Longitudinal growth-curve modeling using hierarchical linear models (HLM)" is then presented as offering "a partial solution to the third variable problem" (p. 312).
The core of the argument appears to be this: "By treating a predictor as a time-varying covariate in the prediction of trajectories, one can rule out the possibility of all time invariant confounds (e.g., relatively stable variables such as socioeconomic status). Specifically, if short-term changes in a predictor predict subsequent short-term changes in achievement, a confounding variable z would have to predict these changes and also be tightly yoked to changes in the predictor over time (i.e., the confound and predictor would have to go up and down together in synchrony over time)" (p. 312).
The authors indeed found self-control to predict GPA longitudinally (and not the reverse), concluding as follows:
"This longitudinal HLM study illustrated an innovative analytic strategy that effectively controlled for all time-invariant third-variable confounds... What our analyses did not rule out, however, is the possibility of an unmeasured time-varying third variable that changes in sync with self-control and causally determines subsequent academic performance" (p. 316; my emphasis added).
I would recommend this article for its excellent exposition on causality with non-experimental designs and for the approach it demonstrates. There are a lot of technical details in the article, as well, and even readers experienced with some of the general techniques used may find themselves having to pause periodically to wrap their minds around each specific analytic decision the authors describe.
Angela Lee Duckworth, Eli Tsukayama, and Henry May have published an article entitled, "Establishing Causality Using Longitudinal Hierarchical Linear Modeling: An Illustration Predicting Achievement From Self-Control" (abstract) in the October 2010 issue of the new journal Social Psychological and Personality Science.
The authors are interested in the personality trait of self-control (which encompasses such abilities as persistence and delay of gratification) and whether it can actually be shown to cause academic achievement (GPA) during the years from fifth to eighth grade. Duckworth and colleagues acknowledge the difficulty early on:
"Manipulating personality in a random-assignment experiment could, in theory, establish its causal role for later outcomes but, alas, personality is not easily manipulated. To our knowledge, no empirical investigation to date has successfully manipulated trait-level self-control and measured subsequent effects on life outcomes" (p. 311).
The authors also carefully review the arguments for why longitudinal predictive models using path-analysis and structural equation modeling (controlling for prior levels of the dependent variable), though an improvement over cross-sectional correlations, are still vulnerable to unmeasured third variables. "Longitudinal growth-curve modeling using hierarchical linear models (HLM)" is then presented as offering "a partial solution to the third variable problem" (p. 312).
The core of the argument appears to be this: "By treating a predictor as a time-varying covariate in the prediction of trajectories, one can rule out the possibility of all time invariant confounds (e.g., relatively stable variables such as socioeconomic status). Specifically, if short-term changes in a predictor predict subsequent short-term changes in achievement, a confounding variable z would have to predict these changes and also be tightly yoked to changes in the predictor over time (i.e., the confound and predictor would have to go up and down together in synchrony over time)" (p. 312).
The authors indeed found self-control to predict GPA longitudinally (and not the reverse), concluding as follows:
"This longitudinal HLM study illustrated an innovative analytic strategy that effectively controlled for all time-invariant third-variable confounds... What our analyses did not rule out, however, is the possibility of an unmeasured time-varying third variable that changes in sync with self-control and causally determines subsequent academic performance" (p. 316; my emphasis added).
I would recommend this article for its excellent exposition on causality with non-experimental designs and for the approach it demonstrates. There are a lot of technical details in the article, as well, and even readers experienced with some of the general techniques used may find themselves having to pause periodically to wrap their minds around each specific analytic decision the authors describe.
Tuesday, September 28, 2010
Psychological Methods Article: Harder et al.
by Alan
Nearly two years ago, we first mentioned propensity scores as a tool for trying to draw as strong a causal inference as possible from nonexperimental designs, between exposure to a "treatment" experience and some later outcome. For example, how might attending private (vs. public) schools affect students' intellectual curiosity? An article by Harder, Stuart, and Anthony in the September 2010 Psychological Methods offers practical advice and concrete examples for conducting propensity-score analyses. Propensity scores involve matching the treatment and control groups on other variables (covariates) that are thought to predict treatment status or have "potential to confound the relationship between the treatment and the outcome" (p. 235). Harder et al. discuss several specific methods for implementing such matching, such as 1:1, 1:k (where one treated participant can be matched to multiple comparison persons), and full matching (where one matched set can include multiple treated individuals and multiple comparison counterparts). The article also discusses software packages for implementing propensity scores, although these do not appear very plentiful at the moment. The article is definitely worth a look by anyone contemplating a correlational or quasi-experimental study that one hopes to frame in causal terms (acknowledging that a full causal claim will be out of reach with such designs).
Nearly two years ago, we first mentioned propensity scores as a tool for trying to draw as strong a causal inference as possible from nonexperimental designs, between exposure to a "treatment" experience and some later outcome. For example, how might attending private (vs. public) schools affect students' intellectual curiosity? An article by Harder, Stuart, and Anthony in the September 2010 Psychological Methods offers practical advice and concrete examples for conducting propensity-score analyses. Propensity scores involve matching the treatment and control groups on other variables (covariates) that are thought to predict treatment status or have "potential to confound the relationship between the treatment and the outcome" (p. 235). Harder et al. discuss several specific methods for implementing such matching, such as 1:1, 1:k (where one treated participant can be matched to multiple comparison persons), and full matching (where one matched set can include multiple treated individuals and multiple comparison counterparts). The article also discusses software packages for implementing propensity scores, although these do not appear very plentiful at the moment. The article is definitely worth a look by anyone contemplating a correlational or quasi-experimental study that one hopes to frame in causal terms (acknowledging that a full causal claim will be out of reach with such designs).
Friday, June 4, 2010
Periodic Message from Judea Pearl (June 2010)
Judea Pearl recently released a message on behalf of the UCLA Causality Blog for interested scholars. It is reprinted below (with light editing, e.g., to make it easier to follow links):
Dear friends in causality,
Below are a few items you might find to be of some interest and possibly some challenge.
1. A new book containing a collection of recent articles on causation, some tutorial in nature, is now available from College Publications (2010).
Title: Heuristics, Probability and Causality
Editors: R. Dechter, H. Geffner and J. Halpern
For table of contents, preface and more information please click here. As you can see, I have had a natural indirect effect on the cover design, but zero controlled direct effect.
2. A symposium on causality and related topics by some of the contributors to Heuristics, Probabilities and Causality was held at UCLA on March 12. Videos of lectures, by C. Hitchcock, S. Greenland, T. Richardson, J. Robins, R. Scheines, J. Tian, Y. Shoham and J. Pearl, can be viewed here. Videos of additional lectures will be posted in the near future.
3. Recent entries on our Causality Blog include:
3.1
An open letter from Judea Pearl to Nancy Cartwright concerning "Causal Pluralism," a topic central to a discussion of her book Hunting Causes, which appeared recently in Economics and Philosophy 26:69-77.(Posted May 31, 2010)
3.2
A lively discussion by T. Richardson, J. Robins and J. Pearl on the structure of the causal hierarchy and the scientific role of untestable counterfactual assumptions.
(Posted May 3 and May 15, 2010)
4. A recent posting on my web-page is a paper titled, "The Mediation Formula: A guide to the assessment of causal pathways in non-linear models," which explains why traditional methods of mediation analysis yield distorted results when applied to discrete data, even when correct parametric models are assumed and all parameters are known precisely. The Mediation Formula circumvents these difficulties.
5. Another posting of potential interest is Technical Report R-364, by T. Kyono (Master Thesis), titled "Commentator: A Front-End User-Interface Module for Graphical and Structural Equation Modeling." It takes a DAG as input and prints: (1) all identifiable direct effects, (2) all identifiable causal effects, (3) all (minimal) sets of admissible covariates, (4) all instrumental variables, and (5) (almost) all testable implications of a model. The source code is available upon request.
6. Finally, I have received inquiries regarding a slide that I used at NYU, in which an instrumental variable poses as an innocent confounder and, upon adjustment, amplifies, rather than reduces confounding bias. The moral of the story was (and is) that "outcome assignment" is safer to model than "treatment assignment." The pertinent paper is R-356 (link).
7. As always, your thoughts are welcome and will surely be put into some good cause when conveyed to other blog readers.
Best,
Judea Pearl
UCLA
Dear friends in causality,
Below are a few items you might find to be of some interest and possibly some challenge.
1. A new book containing a collection of recent articles on causation, some tutorial in nature, is now available from College Publications (2010).
Title: Heuristics, Probability and Causality
Editors: R. Dechter, H. Geffner and J. Halpern
For table of contents, preface and more information please click here. As you can see, I have had a natural indirect effect on the cover design, but zero controlled direct effect.
2. A symposium on causality and related topics by some of the contributors to Heuristics, Probabilities and Causality was held at UCLA on March 12. Videos of lectures, by C. Hitchcock, S. Greenland, T. Richardson, J. Robins, R. Scheines, J. Tian, Y. Shoham and J. Pearl, can be viewed here. Videos of additional lectures will be posted in the near future.
3. Recent entries on our Causality Blog include:
3.1
An open letter from Judea Pearl to Nancy Cartwright concerning "Causal Pluralism," a topic central to a discussion of her book Hunting Causes, which appeared recently in Economics and Philosophy 26:69-77.(Posted May 31, 2010)
3.2
A lively discussion by T. Richardson, J. Robins and J. Pearl on the structure of the causal hierarchy and the scientific role of untestable counterfactual assumptions.
(Posted May 3 and May 15, 2010)
4. A recent posting on my web-page is a paper titled, "The Mediation Formula: A guide to the assessment of causal pathways in non-linear models," which explains why traditional methods of mediation analysis yield distorted results when applied to discrete data, even when correct parametric models are assumed and all parameters are known precisely. The Mediation Formula circumvents these difficulties.
5. Another posting of potential interest is Technical Report R-364, by T. Kyono (Master Thesis), titled "Commentator: A Front-End User-Interface Module for Graphical and Structural Equation Modeling." It takes a DAG as input and prints: (1) all identifiable direct effects, (2) all identifiable causal effects, (3) all (minimal) sets of admissible covariates, (4) all instrumental variables, and (5) (almost) all testable implications of a model. The source code is available upon request.
6. Finally, I have received inquiries regarding a slide that I used at NYU, in which an instrumental variable poses as an innocent confounder and, upon adjustment, amplifies, rather than reduces confounding bias. The moral of the story was (and is) that "outcome assignment" is safer to model than "treatment assignment." The pertinent paper is R-356 (link).
7. As always, your thoughts are welcome and will surely be put into some good cause when conveyed to other blog readers.
Best,
Judea Pearl
UCLA
Monday, May 17, 2010
Special Series in March 2010 Psychological Methods on Quasi-Experiments and Causation
by Alan
Being able to demonstrate causality in social science research is a daunting task. Even in a tightly controlled one-hour laboratory experiment, there are potential threats such as experimenter bias or demand characteristics. In long-term field studies, even with random assignment, there is a lot of time for things to go wrong, namely the threats to internal validity enumerated by Campbell (with colleagues Stanley and Cook) such as history effects, attrition, or diffusion and imitation of treatments (here and here). Finally, in quasi-experiments, which lack the crucial element of random assignment, one knows from the start that a full causal inference is out of reach.
Where Cook and Rubin differ is in their main techniques for trying to get as much probative evidence as possible out of a given research design (experimental, quasi-experimental, or correlational/observational). As Shadish notes, with some qualification, "[Rubin] has focused more on analysis, whereas [Campbell] has focused more on design" (p. 11). West and Thoemmes phrase the contrast as follows: "Campbell's perspective emphasizes the prevention of threats to internal validity rather than their correction" (p. 22).
Campbell's approach, in essence, seems to be examining ahead of time where the flaws in a research design might lie, and then shoring up the design as best as possible. The latter might be done via steps such as pre-testing, extra control groups, and extra dependent variables (any of which might be affected by an artifactual phenomenon, but only one of which should be affected by the intervention). West and Thoemmes describe a previous quasi-experimental study by Reynolds and West (1987), who sought to test the effectiveness of an intervention to help convenience stores sell lottery tickets. This example, which involved matched control groups because store managers would not accept random assignment, was very clear and came with accompanying graphics to illustrate interpretation of the findings.
Rubin's approach relies on quantitative operations to assemble comparable experimental and control groups, when original group membership was not determined randomly. Propensity scores appear to be the main weapon in Rubin's arsenal. West and Thoemmes also walk the reader through an example of propensity scores, using 2008 research by Wu, West, and Hughes on retention in first grade and later academic achievement.
Even with my background teaching research methods for 13 years and working on projects using quasi-experimentation and propensity scores, I still found the articles highly educational. For example, the idea that Campbell's internal-validity-threats framework could be taken beyond the simple presence or absence of threats to possible quantification of the magnitude of the threats' effect was new to me. I also found discussion of the continuing controversy over matching to be highly informative. (As part of the extended commentary, Rubin writes in his piece about being "flabbergasted" upon hearing Campbell's denunciation of matching in a face-to-face meeting between the two in the early 1970s [p. 39].)
I enthusiastically recommend the Psych Methods special series as an aid to non-experimental research and as assigned reading for graduate-level methodology courses.
Sunday, November 1, 2009
Child Development Article on Spanking and Early Child Development
by Alan
The September/October 2009 issue of Child Development includes an article entitled, "Correlates and Consequences of Spanking and Verbal Punishment for Low-Income White, African American, and Mexican American Toddlers." The article was authored by Duke University's Lisa Berlin and a long list of co-authors from multiple institutions; the eight authors of the article were themselves representing an even larger set of investigators who formed the Early Head Start Research Consortium. The abstract of the article is available here. I would like to thank Jonathan Mueller for bringing the article to my attention and suggesting I comment upon it.
The Berlin et al. article is an example of a longitudinal/panel study, from which causal inference is potentially very good, but never complete (see earlier posting on this topic). Given the emotional reaction many people have to spanking and related issues, however, any article of this type is bound to receive great scrutiny.
The study drew a sample of 2,573 children and their primary caregivers (99% mothers) from 17 sites nationally. Families were assessed when the children were 1, 2, and 3 years of age, with key study variables including parental spanking and verbal punishment, and child fussiness (EASI II temperamental emotionality; age 1 only), aggression (CBCL), and mental development (Bayley scores). The latter two child outcomes were measured only at ages 2 and 3. The authors addressed causality issues in the Introduction, on pp. 1405-1406:
In keeping with transactional theories of child development... another question requiring further study concerns the direction of effects. In particular, to what extent do parental discipline strategies drive child outcomes, to what extent are these parenting strategies elicited by particular child behaviors, and to what extent are both causal mechanisms operative? ... As recommended by Gershoff and Bitensky, cross-lagged path models that simultaneously estimate effects from parental discipline strategies to child behaviors and vice versa are critical to disentangling such issues.
Some of the key statistically significant findings from the regression analyses were as follows:
Greater age-1 child fussiness was associated with greater age-1 and -2 spanking and verbal punishment;
Greater age-1 spanking was associated with greater age-2 child aggression; and
Greater age-1 spanking was associated with lower age-3 child mental development.
Two of the criteria for causality -- presence of statistical association and time-ordering -- are clearly met for the above relationships. The third and final criterion is that all possible "third variables" (e.g., something that might cause both spanking and child aggression) are ruled out. No study can ever rule out all possible third variables, so a more realistic question is whether the investigators ruled out the most plausible contenders. Quoting from the Table 5 caption, Berlin et al report controlling for "Early Head Start program participation, maternal race/ethnicity, age, and education, maternal depression at age 1, family income and structure, and child sex." It also appears from Table 5 and Figure 1 that the significant path from age-1 spanking to age-2 child aggression was obtained while controlling for age-1 child fussiness (i.e., an age-1 fussiness to age-2 aggression path was also included in the model, and was significantly positive).
The above set of control variables seems fairly comprehensive, but observers can usually suggest more (some more plausible than others). A skeptic might note that the study design was not "genetically informed," in other words, not able to examine rigorously whether, for example, genetic tendencies toward irritability that may have been shared between mother and child may have contributed to the obtained relationships (a phenomenon known as passive gene-environment correlation). Controlling for child fussiness probably helps a little bit in overcoming this objection, although it would have been good to control for symptoms of other forms of maternal psychopathology besides depression.
It is important to add, however, that the authors appear to have gone to great pains to examine the possible strengths and weaknesses of their study, in order to present an honest appraisal of the findings. Specifically, they undertook a number of supplementary analyses to (potentially) qualify the scope of their conclusions, including tests for whether the basic results held up equally well across different racial/ethnic groups (i.e., moderation) and whether removing the most "severe" spankers affected the results. They also discussed what they perceived as limitations to their own study, such as the self-report measure of spanking not containing a specific definition of the act, and acknowledged that the results suggesting an effect of spanking were of "small," though significantly non-zero, magnitude.
Berlin et al. summarize their study as follows:
[The findings] support the conclusion that spanking during toddlerhood can have negative consequences for toddlers' cognitive as well as socioemotional functioning (p. 1417).
I find this conclusion appropriate. As noted, the authors' research design approaches causality about as well as is possible with a non-experimental study, and the use of "can" as a qualifier in the preceding quote conveys the necessary caution.
The September/October 2009 issue of Child Development includes an article entitled, "Correlates and Consequences of Spanking and Verbal Punishment for Low-Income White, African American, and Mexican American Toddlers." The article was authored by Duke University's Lisa Berlin and a long list of co-authors from multiple institutions; the eight authors of the article were themselves representing an even larger set of investigators who formed the Early Head Start Research Consortium. The abstract of the article is available here. I would like to thank Jonathan Mueller for bringing the article to my attention and suggesting I comment upon it.
The Berlin et al. article is an example of a longitudinal/panel study, from which causal inference is potentially very good, but never complete (see earlier posting on this topic). Given the emotional reaction many people have to spanking and related issues, however, any article of this type is bound to receive great scrutiny.
The study drew a sample of 2,573 children and their primary caregivers (99% mothers) from 17 sites nationally. Families were assessed when the children were 1, 2, and 3 years of age, with key study variables including parental spanking and verbal punishment, and child fussiness (EASI II temperamental emotionality; age 1 only), aggression (CBCL), and mental development (Bayley scores). The latter two child outcomes were measured only at ages 2 and 3. The authors addressed causality issues in the Introduction, on pp. 1405-1406:
In keeping with transactional theories of child development... another question requiring further study concerns the direction of effects. In particular, to what extent do parental discipline strategies drive child outcomes, to what extent are these parenting strategies elicited by particular child behaviors, and to what extent are both causal mechanisms operative? ... As recommended by Gershoff and Bitensky, cross-lagged path models that simultaneously estimate effects from parental discipline strategies to child behaviors and vice versa are critical to disentangling such issues.
Some of the key statistically significant findings from the regression analyses were as follows:
Greater age-1 child fussiness was associated with greater age-1 and -2 spanking and verbal punishment;
Greater age-1 spanking was associated with greater age-2 child aggression; and
Greater age-1 spanking was associated with lower age-3 child mental development.
Two of the criteria for causality -- presence of statistical association and time-ordering -- are clearly met for the above relationships. The third and final criterion is that all possible "third variables" (e.g., something that might cause both spanking and child aggression) are ruled out. No study can ever rule out all possible third variables, so a more realistic question is whether the investigators ruled out the most plausible contenders. Quoting from the Table 5 caption, Berlin et al report controlling for "Early Head Start program participation, maternal race/ethnicity, age, and education, maternal depression at age 1, family income and structure, and child sex." It also appears from Table 5 and Figure 1 that the significant path from age-1 spanking to age-2 child aggression was obtained while controlling for age-1 child fussiness (i.e., an age-1 fussiness to age-2 aggression path was also included in the model, and was significantly positive).
The above set of control variables seems fairly comprehensive, but observers can usually suggest more (some more plausible than others). A skeptic might note that the study design was not "genetically informed," in other words, not able to examine rigorously whether, for example, genetic tendencies toward irritability that may have been shared between mother and child may have contributed to the obtained relationships (a phenomenon known as passive gene-environment correlation). Controlling for child fussiness probably helps a little bit in overcoming this objection, although it would have been good to control for symptoms of other forms of maternal psychopathology besides depression.
It is important to add, however, that the authors appear to have gone to great pains to examine the possible strengths and weaknesses of their study, in order to present an honest appraisal of the findings. Specifically, they undertook a number of supplementary analyses to (potentially) qualify the scope of their conclusions, including tests for whether the basic results held up equally well across different racial/ethnic groups (i.e., moderation) and whether removing the most "severe" spankers affected the results. They also discussed what they perceived as limitations to their own study, such as the self-report measure of spanking not containing a specific definition of the act, and acknowledged that the results suggesting an effect of spanking were of "small," though significantly non-zero, magnitude.
Berlin et al. summarize their study as follows:
[The findings] support the conclusion that spanking during toddlerhood can have negative consequences for toddlers' cognitive as well as socioemotional functioning (p. 1417).
I find this conclusion appropriate. As noted, the authors' research design approaches causality about as well as is possible with a non-experimental study, and the use of "can" as a qualifier in the preceding quote conveys the necessary caution.
UPDATE (May 17, 2010): Robert Larzelere, a faculty member at Oklahoma State University whose interests include causal inference from longitudinal surveys, e-mailed me a few weeks ago that he and his colleagues had published some articles on spanking and alternative sources of discipline. Here's a link to his list of publications, from which interested readers can see the kinds of issues Dr. Larzelere has raised regarding causal inference and parental discipline.
Thursday, October 22, 2009
Conservative Religiosity and Teen Birthrates in States
by Alan
Nick Barrowman, who operates the blog "Log base 2," examines a study reported on MSNBC.com showing a strong correlation (.72) between states' teenage birthrates and conservative religiosity (the latter obtained via public opinion polling in each state). Both the MSNBC article and the original scientific publication discuss how caution is warranted in interpreting the findings, both in terms of correlation and causality, and the ecological fallacy. Indeed, Barrowman writes that, "My goal here has not been to criticize the authors of this study, nor the media." Instead, his concern appears to be that the general public may draw improper conclusions from the study. All in all, however, I would say the scientific article -- and media and blog coverage thereof -- have done a public service by promoting an exchange of ideas about how the study's findings might reasonably be explained.
Nick Barrowman, who operates the blog "Log base 2," examines a study reported on MSNBC.com showing a strong correlation (.72) between states' teenage birthrates and conservative religiosity (the latter obtained via public opinion polling in each state). Both the MSNBC article and the original scientific publication discuss how caution is warranted in interpreting the findings, both in terms of correlation and causality, and the ecological fallacy. Indeed, Barrowman writes that, "My goal here has not been to criticize the authors of this study, nor the media." Instead, his concern appears to be that the general public may draw improper conclusions from the study. All in all, however, I would say the scientific article -- and media and blog coverage thereof -- have done a public service by promoting an exchange of ideas about how the study's findings might reasonably be explained.
Saturday, July 18, 2009
Causality in Parent-Child Dynamics
by Alan
I recently finished reading a new book by Johns Hopkins University sociologist Andrew Cherlin entitled The Marriage-Go-Round: The State of Marriage and the Family in America Today. As the title implies, the U.S. has a lot of couple and family turnover. Though Americans' high rates of marriage and divorce are well-known, there's a third element, of which I wasn't really familiar. Namely, Americans also have a high rate of re-partnering after the break-up of marital and non-marital couples. An illustrative statistic Cherlin cites is the percentage of women in different countries who have "three or more live-in partners (married or cohabiting) by age thirty-five" (p. 19). In the U.S., it's 10%, whereas in other English-speaking nations (those in Europe, as well as Canada, Australia, and New Zealand), none was higher than 4.5%.
Cherlin notes further that, "Children who experiences a series of transitions appear to have more difficulties than children raised in stable two-parent families and perhaps even more than children raised in stable lone-parent families" (p. 20). He acknowledges, however, that, "we cannot be sure that experiencing parents and partners moving in and out of the house actually causes the difficulties researchers have found in children. Some aspects of the parents' personalities or abilities could affect both the stability of their partnerships and their children's behavior" (pp. 20-21). The possibility that a genetic-based factor, present in the parent and passed to the children, could cause both the parent's relationship difficulties and the children's behavior problems is also acknowledged.
Cherlin discusses two research approaches that attempt to get around these interpretational difficulties. Regarding the genetic issue:
A way to test this possibility is to compare the adjustment of biological children, who share their parents' genes, with adopted children, who do not. If having genes in common is the root cause of the difficulties we see in families of divorce, we would expect that biological children of divorced parents would show more problems after a parental divorce than would adopted children. But that's not what researchers find... (p. 21).
I recently finished reading a new book by Johns Hopkins University sociologist Andrew Cherlin entitled The Marriage-Go-Round: The State of Marriage and the Family in America Today. As the title implies, the U.S. has a lot of couple and family turnover. Though Americans' high rates of marriage and divorce are well-known, there's a third element, of which I wasn't really familiar. Namely, Americans also have a high rate of re-partnering after the break-up of marital and non-marital couples. An illustrative statistic Cherlin cites is the percentage of women in different countries who have "three or more live-in partners (married or cohabiting) by age thirty-five" (p. 19). In the U.S., it's 10%, whereas in other English-speaking nations (those in Europe, as well as Canada, Australia, and New Zealand), none was higher than 4.5%.
Cherlin notes further that, "Children who experiences a series of transitions appear to have more difficulties than children raised in stable two-parent families and perhaps even more than children raised in stable lone-parent families" (p. 20). He acknowledges, however, that, "we cannot be sure that experiencing parents and partners moving in and out of the house actually causes the difficulties researchers have found in children. Some aspects of the parents' personalities or abilities could affect both the stability of their partnerships and their children's behavior" (pp. 20-21). The possibility that a genetic-based factor, present in the parent and passed to the children, could cause both the parent's relationship difficulties and the children's behavior problems is also acknowledged.
Cherlin discusses two research approaches that attempt to get around these interpretational difficulties. Regarding the genetic issue:
A way to test this possibility is to compare the adjustment of biological children, who share their parents' genes, with adopted children, who do not. If having genes in common is the root cause of the difficulties we see in families of divorce, we would expect that biological children of divorced parents would show more problems after a parental divorce than would adopted children. But that's not what researchers find... (p. 21).
In the endnotes (pp. 216-217), Cherlin adds the following:
Paula Fomby and I looked at this question another way. If what's happening is merely that parents are passing along traits that lead to difficulties, then, we reasoned, children who are acting out or delinquent should be more likely to have parents who acted out or were delinquent when they were children... We examined the records of a twenty-year national study that followed women beginning when they were teenagers. Most of the women became mothers during the study. We found that even after we took into account whether the mothers had, when they were teenagers, used drugs, shoplifted, stolen something, or had early sexual intercourse, their children still had more behavior problems and admitted to more delinquency if their mothers had had more partners. Our study suggests that experiencing a series of partnerships may be, at least in part, a true cause of children's difficulties.
These findings were said to hold for white families, but not for black families. With non-experimental research, alternative explanations for findings are virtually always present. The research described by Cherlin is noteworthy, in my mind, for the creative research designs used in an attempt to rule out some of the leading alternative interpretations.Thursday, July 9, 2009
SEMNET Message from Judea Pearl
The following is a message from Judea Pearl to the SEMNET (Structural Equation Modeling) discussion list (conveyed by Dr. Pearl's colleague Stephen Sivo). Given Dr. Pearl's apparent intent to have these announcements publicized to a wide audience, I (Alan) have reprinted it below (lightly edited for apparent typos and ease of accessing links)...
Dear Colleagues in Causality,
Below, a few items that I thought would be of interest to researchers active in causal reasoning.
1. A new article, authored by Ilya Shpitser and myself is now posted on the UCLA Causality-Blog (see also here). It offers a solution to the problem of evaluating "Effects of Treatment on the Treated (ETT)." The problem is of theoretical interest because ETT, despite its blatant counterfactual character (e.g., "I just took an aspirin, perhaps I shouldn't have?"), can be evaluated from experimental studies in many, though not all, cases. Characterizing those cases illuminates therefore the empirical content of counterfactuals.
2. Many of you have commented on my article "Myth, Confusion and Science in Causal Analysis" (inspired by Don Rubin), a revised version of which is now posted on our website here . I would like to encourage a blog-discussion on the main points raised there. For example:
2.1. Whether graphical methods are in some way "less principled" than other methods of analysis.
2.2. Whether confounding bias can only decrease by conditioning on a new covariate.
2.3 Whether the M-bias, when it occurs, is merely a mathematical curiosity, unworthy of researchers' attention.
2.4. Whether Bayesianism instructs us to condition on all available measurements.
If you feel strongly about defending any of these claims (which seem to be
still simmering in certain circles, see video), the Causality-Blog can be an effective arena for airing them in an open discussion. Requests for anonymity will be honored.
3. Forbes magazine ran an issue on artificial intelligence last week, to which I contributed a popular article on progress in causal analysis (from my humble perspective, of course). Comments are welcome.
4. I have volunteered to give a tutorial at the JSM meeting (Washington, DC, August 5, 2009, 2-4 pm) on "Causal Analysis in Statistics: A Gentle Introduction." If any of your colleagues or students could benefit from such tutorial, I promise to be truly gentle.
5. Just before the tutorial, at 12 noon, there will be a book-signing gathering at the Cambridge University Press booth, where I will be signing copies of the 2nd Edition of Causality (and will engage in gossip and debates about where causality is heading).
Best wishes, and May clarity shine over causality land,
Judea Pearl
Note that all of Dr. Pearl's requests for blog discussion pertain to his blog at UCLA. I have thus turned off the comments option here.
Dear Colleagues in Causality,
Below, a few items that I thought would be of interest to researchers active in causal reasoning.
1. A new article, authored by Ilya Shpitser and myself is now posted on the UCLA Causality-Blog (see also here). It offers a solution to the problem of evaluating "Effects of Treatment on the Treated (ETT)." The problem is of theoretical interest because ETT, despite its blatant counterfactual character (e.g., "I just took an aspirin, perhaps I shouldn't have?"), can be evaluated from experimental studies in many, though not all, cases. Characterizing those cases illuminates therefore the empirical content of counterfactuals.
2. Many of you have commented on my article "Myth, Confusion and Science in Causal Analysis" (inspired by Don Rubin), a revised version of which is now posted on our website here . I would like to encourage a blog-discussion on the main points raised there. For example:
2.1. Whether graphical methods are in some way "less principled" than other methods of analysis.
2.2. Whether confounding bias can only decrease by conditioning on a new covariate.
2.3 Whether the M-bias, when it occurs, is merely a mathematical curiosity, unworthy of researchers' attention.
2.4. Whether Bayesianism instructs us to condition on all available measurements.
If you feel strongly about defending any of these claims (which seem to be
still simmering in certain circles, see video), the Causality-Blog can be an effective arena for airing them in an open discussion. Requests for anonymity will be honored.
3. Forbes magazine ran an issue on artificial intelligence last week, to which I contributed a popular article on progress in causal analysis (from my humble perspective, of course). Comments are welcome.
4. I have volunteered to give a tutorial at the JSM meeting (Washington, DC, August 5, 2009, 2-4 pm) on "Causal Analysis in Statistics: A Gentle Introduction." If any of your colleagues or students could benefit from such tutorial, I promise to be truly gentle.
5. Just before the tutorial, at 12 noon, there will be a book-signing gathering at the Cambridge University Press booth, where I will be signing copies of the 2nd Edition of Causality (and will engage in gossip and debates about where causality is heading).
Best wishes, and May clarity shine over causality land,
Judea Pearl
Note that all of Dr. Pearl's requests for blog discussion pertain to his blog at UCLA. I have thus turned off the comments option here.
Monday, June 15, 2009
Moderate Alcohol Consumption and Heart Health
by Alan
Today's New York Times has an article on the link between moderate alcohol consumption and lower heart disease, and whether health officials should actively recommend a daily drink or two for the public. At a correlational level, the alcohol-heart health relationship appears well established. As the piece in the Times delves into, the lack of true experimentation (in this case via Randomized Clinical Trials) gives skeptics something to hang their hats on. The following are some key excerpts of the article:
For some scientists, the question will not go away. No study, these critics say, has ever proved a causal relationship between moderate drinking and lower risk of death — only that the two often go together. It may be that moderate drinking is just something healthy people tend to do, not something that makes people healthy.
“The moderate drinkers tend to do everything right — they exercise, they don’t smoke, they eat right and they drink moderately,” said Kaye Middleton Fillmore, a retired sociologist from the University of California, San Francisco, who has criticized the research. “It’s very hard to disentangle all of that, and that’s a real problem.” ...
“The bottom line is there has not been a single study done on moderate alcohol consumption and mortality outcomes that is a ‘gold standard’ kind of study — the kind of randomized controlled clinical trial that we would be required to have in order to approve a new pharmaceutical agent in this country,” said Dr. Tim Naimi, an epidemiologist with the Centers for Disease Control and Prevention.
The article goes on to discuss how clinical trials might be designed, but also their potential ethical and logistical difficulties.
Today's New York Times has an article on the link between moderate alcohol consumption and lower heart disease, and whether health officials should actively recommend a daily drink or two for the public. At a correlational level, the alcohol-heart health relationship appears well established. As the piece in the Times delves into, the lack of true experimentation (in this case via Randomized Clinical Trials) gives skeptics something to hang their hats on. The following are some key excerpts of the article:
For some scientists, the question will not go away. No study, these critics say, has ever proved a causal relationship between moderate drinking and lower risk of death — only that the two often go together. It may be that moderate drinking is just something healthy people tend to do, not something that makes people healthy.
“The moderate drinkers tend to do everything right — they exercise, they don’t smoke, they eat right and they drink moderately,” said Kaye Middleton Fillmore, a retired sociologist from the University of California, San Francisco, who has criticized the research. “It’s very hard to disentangle all of that, and that’s a real problem.” ...
“The bottom line is there has not been a single study done on moderate alcohol consumption and mortality outcomes that is a ‘gold standard’ kind of study — the kind of randomized controlled clinical trial that we would be required to have in order to approve a new pharmaceutical agent in this country,” said Dr. Tim Naimi, an epidemiologist with the Centers for Disease Control and Prevention.
The article goes on to discuss how clinical trials might be designed, but also their potential ethical and logistical difficulties.
Tuesday, March 10, 2009
New Book: The Numbers Game
by Alan
Journalist Michael Blastland and economist Andrew Dilnot host a radio show on Great Britain's BBC4 called More or Less, which attempts to engage the public on issues of data quality, statistics, and conclusion-drawing with regard to numbers reported by the media, politicians, etc. Blastland and Dilnot have also put their ideas into book form, with the recent U.S. release of The Numbers Game: The Commonsense Guide to Understanding Numbers in the News, in Politics, and in Life (previously published in the UK as The Tiger That Isn't).
I've just read The Numbers Game and, for purposes of the present blog, Chapter 12 on "Causation" is most relevant. The book's chapters are generally around 10-20 pages each, with the Causation chapter toward the shorter end. Much of this chapter presents standard material, such as a set of correlational scenarios that might tempt a reader to draw causal conclusions but ultimately turn out to be more plausibly explained by third variables. Beyond this, however, I feel the authors provide some useful insights:
What seems often to determine how easily we spot causation/correlation errors is how fast a better explanation comes to mind: thinking of decent alternatives slows conclusions and sows skepticism (p. 186).
AND
Restlessness for the true cause is a constructive habit, an insurance against gullibility. And though correlation does not prove causation, it is often a good hint, but a hint to start asking questions, not to settle for easy answers (191-192)
[As somewhat of an aside, the first of the two above statements bears some similarity in my mind to the late causal-attribution theorist Hal Kelley's concept of discounting.]
I would urge teachers of undergraduate research methods to consider using The Numbers Game (or specific chapters therein) to supplement their main textbooks. The writing is lively and the examples should help students grasp key concepts.
Journalist Michael Blastland and economist Andrew Dilnot host a radio show on Great Britain's BBC4 called More or Less, which attempts to engage the public on issues of data quality, statistics, and conclusion-drawing with regard to numbers reported by the media, politicians, etc. Blastland and Dilnot have also put their ideas into book form, with the recent U.S. release of The Numbers Game: The Commonsense Guide to Understanding Numbers in the News, in Politics, and in Life (previously published in the UK as The Tiger That Isn't).
I've just read The Numbers Game and, for purposes of the present blog, Chapter 12 on "Causation" is most relevant. The book's chapters are generally around 10-20 pages each, with the Causation chapter toward the shorter end. Much of this chapter presents standard material, such as a set of correlational scenarios that might tempt a reader to draw causal conclusions but ultimately turn out to be more plausibly explained by third variables. Beyond this, however, I feel the authors provide some useful insights:
What seems often to determine how easily we spot causation/correlation errors is how fast a better explanation comes to mind: thinking of decent alternatives slows conclusions and sows skepticism (p. 186).
AND
Restlessness for the true cause is a constructive habit, an insurance against gullibility. And though correlation does not prove causation, it is often a good hint, but a hint to start asking questions, not to settle for easy answers (191-192)
[As somewhat of an aside, the first of the two above statements bears some similarity in my mind to the late causal-attribution theorist Hal Kelley's concept of discounting.]
I would urge teachers of undergraduate research methods to consider using The Numbers Game (or specific chapters therein) to supplement their main textbooks. The writing is lively and the examples should help students grasp key concepts.
Sunday, December 28, 2008
Psychological Methods Article: Schafer & Kang
by Alan
The December 2008 issue of Psychological Methods includes an article by Joseph Schafer and Joseph Kang, entitled, "Average Causal Effects From Nonrandomized Studies: A Practical Guide and Simulated Example" (abstract).
The article addresses the seemingly age-old issue of how best to approximate a causal inference when participants' levels of the purported "causal" variable have not been assigned at random. Although the article contains a fair amount of jargon and technical formulas, the key foundation appears to boil down to the following quote:
In a typical observational study... it is unlikely that [treatment condition] will be independent of [individuals' outcome scores]. The treatments may have been selected by the individuals themselves, for reasons that are possibly related to the outcomes. With observational data, a good estimate of the [average causal effect] will make use of the covariates... to help account for this dependence (p. 281).
Via a large simulation study, Schafer and Kang compare nine different approaches for covariate-based adjustment, including analysis of covariance (ANCOVA), regression, matching, propensity scores, and weighting schemes. Near the end of the article, the authors present a section entitled "Lessons Learned," containing practical recommendations.
As social-science research articles continue to use increasingly sophisticated statistical and analytical methods, Schafer and Kang's article should be a useful resource for researchers looking to remain current with state-of-the-art approaches.
The December 2008 issue of Psychological Methods includes an article by Joseph Schafer and Joseph Kang, entitled, "Average Causal Effects From Nonrandomized Studies: A Practical Guide and Simulated Example" (abstract).
The article addresses the seemingly age-old issue of how best to approximate a causal inference when participants' levels of the purported "causal" variable have not been assigned at random. Although the article contains a fair amount of jargon and technical formulas, the key foundation appears to boil down to the following quote:
In a typical observational study... it is unlikely that [treatment condition] will be independent of [individuals' outcome scores]. The treatments may have been selected by the individuals themselves, for reasons that are possibly related to the outcomes. With observational data, a good estimate of the [average causal effect] will make use of the covariates... to help account for this dependence (p. 281).
Via a large simulation study, Schafer and Kang compare nine different approaches for covariate-based adjustment, including analysis of covariance (ANCOVA), regression, matching, propensity scores, and weighting schemes. Near the end of the article, the authors present a section entitled "Lessons Learned," containing practical recommendations.
As social-science research articles continue to use increasingly sophisticated statistical and analytical methods, Schafer and Kang's article should be a useful resource for researchers looking to remain current with state-of-the-art approaches.
Sunday, December 21, 2008
Do Vitamin Supplements Prevent Disease?
by Alan
Today's Los Angeles Times has an article on the failure of several recent randomized clinical trials to show benefits of taking vitamin supplements for disease prevention. The article provides some good explanation for the layperson on the differences between experimental (i.e., RCT) and observational (i.e., correlational) studies for being able to make causal inferences, in my view. It also discusses some problems, specific to vitamin research, that complicate the interpretation of experimental/RCT studies, even though they're designed to hold extraneous factors constant.
First, here are some examples of the article's discussion of experimental vs. observational research:
Randomized clinical trials are designed to test one factor at a time, but vitamins and minerals consumed as part of a healthy diet work in concert with each other.
"You don't eat a food that just has beta carotene in it," said Dr. Mary L. Hardy, medical director of the Simms/Mann UCLA Center for Integrative Oncology...
Also:
"The observational studies that originally linked vitamins to better health may have been biased because people who take supplements are often healthier overall than people who don't.
"They tend to be more physically active, better educated, eat better diets, and tend not to be smokers," [the NIH's Paul] Coates said. "So you can't say for certain it's your item of interest that causes" the health benefit.
The issue that's specific to vitamin studies is as follows, focusing on the need for a "placebo" or "control" group that does not receive the treatment being tested:
For several reasons, researchers say, vitamins don't lend themselves to randomized controlled trials. Chief among them is that there is no true placebo group when it comes to vitamins and minerals, because everyone gets some in their diet."
For drugs, someone either has [the anti-impotence drug] Cialis in their system or he doesn't have Cialis," said Paul Coates, director of the NIH Office of Dietary Supplements in Bethesda, Md. But with vitamins, "there's a baseline exposure that needs to be taken into account. It makes the challenge of seeing an improvement more difficult."
My (Alan's) initial reaction to this statement was that extraneous exposure to vitamins from food consumption should not necessarily compromise researchers' ability to make a causal inference about the vitamin supplements. The experimental group would be ingesting vitamin pills plus vitamins in food, whereas the control group would be ingesting placebo (sugar) pills plus vitamins in food. Assuming the food intake to be similar in the experimental and control groups -- which is what random assignment to groups is supposed to buy us -- then the only difference between the groups should be the vitamin vs. placebo pills.
However, there's yet another complication:
A vitamin's benefit may become apparent only if people aren't getting enough of it. That could explain why vitamin D has been linked to reduced rates of heart disease, cancer and diabetes.
"Most people are vitamin-D-deficient, and that's not true for vitamin E," [the USDA's Jeffrey] Blumberg said.
Today's Los Angeles Times has an article on the failure of several recent randomized clinical trials to show benefits of taking vitamin supplements for disease prevention. The article provides some good explanation for the layperson on the differences between experimental (i.e., RCT) and observational (i.e., correlational) studies for being able to make causal inferences, in my view. It also discusses some problems, specific to vitamin research, that complicate the interpretation of experimental/RCT studies, even though they're designed to hold extraneous factors constant.
First, here are some examples of the article's discussion of experimental vs. observational research:
Randomized clinical trials are designed to test one factor at a time, but vitamins and minerals consumed as part of a healthy diet work in concert with each other.
"You don't eat a food that just has beta carotene in it," said Dr. Mary L. Hardy, medical director of the Simms/Mann UCLA Center for Integrative Oncology...
Also:
"The observational studies that originally linked vitamins to better health may have been biased because people who take supplements are often healthier overall than people who don't.
"They tend to be more physically active, better educated, eat better diets, and tend not to be smokers," [the NIH's Paul] Coates said. "So you can't say for certain it's your item of interest that causes" the health benefit.
The issue that's specific to vitamin studies is as follows, focusing on the need for a "placebo" or "control" group that does not receive the treatment being tested:
For several reasons, researchers say, vitamins don't lend themselves to randomized controlled trials. Chief among them is that there is no true placebo group when it comes to vitamins and minerals, because everyone gets some in their diet."
For drugs, someone either has [the anti-impotence drug] Cialis in their system or he doesn't have Cialis," said Paul Coates, director of the NIH Office of Dietary Supplements in Bethesda, Md. But with vitamins, "there's a baseline exposure that needs to be taken into account. It makes the challenge of seeing an improvement more difficult."
My (Alan's) initial reaction to this statement was that extraneous exposure to vitamins from food consumption should not necessarily compromise researchers' ability to make a causal inference about the vitamin supplements. The experimental group would be ingesting vitamin pills plus vitamins in food, whereas the control group would be ingesting placebo (sugar) pills plus vitamins in food. Assuming the food intake to be similar in the experimental and control groups -- which is what random assignment to groups is supposed to buy us -- then the only difference between the groups should be the vitamin vs. placebo pills.
However, there's yet another complication:
A vitamin's benefit may become apparent only if people aren't getting enough of it. That could explain why vitamin D has been linked to reduced rates of heart disease, cancer and diabetes.
"Most people are vitamin-D-deficient, and that's not true for vitamin E," [the USDA's Jeffrey] Blumberg said.
Wednesday, December 3, 2008
Call for Papers -- Edited Volume
The following Call for Papers appeared on the SEMNET (Structural Equation Modeling) discussion list:
CAUSALITY IN THE SCIENCES
A volume of papers on causality across the sciences edited by Phyllis McKayIllari, Federica Russo and Jon Williamson under contract with Oxford University Press.
http://www.kent.ac.uk/secl/philosophy/jw/2009/cits/
This book will contain original research papers that deal with causality and causal inference in the various sciences and with general questions concerning the relationship between causality, probability and mechanisms. Some chapters will be invited contributions; others will be submitted to a call for papers. All papers will be subject to a reviewing process.
TIMETABLE
1st July 2009: deadline for submission of full papers for publication to be emailed to Phyllis McKay Illari (P.McKay@kent.ac.uk) or Federica Russo (F.Russo@kent.ac.uk)
1st November 2009: notification of acceptance of papers for publication.
1st December 2009: deadline for final version of papers accepted for publication.
THE VOLUME
The volume will run to about 600 pages and will be subdivided into the following parts:
Introduction
Health Sciences
Social Sciences
Natural Sciences
Psychology and Neurosciences
Computer science and statistics
Causality, probability and mechanisms
ORGANISATION
This volume is organised by the Centre for Reasoning at the University of Kent. It is associated with the Causality in the Sciences series of conferences, and with the research project Mechanisms and Causality funded by the Leverhulme Trust.
CAUSALITY IN THE SCIENCES
A volume of papers on causality across the sciences edited by Phyllis McKayIllari, Federica Russo and Jon Williamson under contract with Oxford University Press.
http://www.kent.ac.uk/secl/philosophy/jw/2009/cits/
This book will contain original research papers that deal with causality and causal inference in the various sciences and with general questions concerning the relationship between causality, probability and mechanisms. Some chapters will be invited contributions; others will be submitted to a call for papers. All papers will be subject to a reviewing process.
TIMETABLE
1st July 2009: deadline for submission of full papers for publication to be emailed to Phyllis McKay Illari (P.McKay@kent.ac.uk) or Federica Russo (F.Russo@kent.ac.uk)
1st November 2009: notification of acceptance of papers for publication.
1st December 2009: deadline for final version of papers accepted for publication.
THE VOLUME
The volume will run to about 600 pages and will be subdivided into the following parts:
Introduction
Health Sciences
Social Sciences
Natural Sciences
Psychology and Neurosciences
Computer science and statistics
Causality, probability and mechanisms
ORGANISATION
This volume is organised by the Centre for Reasoning at the University of Kent. It is associated with the Causality in the Sciences series of conferences, and with the research project Mechanisms and Causality funded by the Leverhulme Trust.
Subscribe to:
Posts (Atom)