The Wall Street Journal posed a challenge to scientists in yesterday's paper: "Is an Economist Qualified To Solve Puzzle of Autism?" Author Mark Whitehouse looked at the controversy stirred up by Cornell economist Michael Waldman's study¹ last October that linked TV viewing to autism. In "Does Television Cause Autism?" Waldman used precipitation records and cable subscriptions as proxies for TV viewing, then performed statistical analysis to correlate television watching with incidence of autism.
Waldman was motivated to study autism by his family's experience with his young son, who was affected by Autism Spectrum Disorder. In response to his son's diagnosis, and in addition to doctor recommended therapies, he curtailed his son's television watching. To his surprise the child recovered completely. However he was unable to engage doctors to study whether TV caused autism, so he studied the connection himself. He found a causal effect in his study and recommended that parents not allow young children to watch TV. As the WSJ article recounts, many researchers don't agree with his conclusion. Scientists, autism researchers especially, were most critical, but economists also questioned his methods. Although his methods weren't unheard of, some economists said the "instrumental-variables technique" was imperfect and others said it tempted economists to study topics they're "not particularly well-trained" to study. Acronym Required wrote a satirical post on the study last October².
Despite the impression given by the Wall Street Journal, Waldman's self-reliant approach to setting a science research agenda is not unprecedented. Other people whose kids are afflicted by autism have also poured personal resources into autism research. A 2005 Wall Street Journal article, "A Hedge-fund Titan Stirs up Research into Autism", mathematician James Simons, who, motivated by his daughter's autism, founded an organization that plans to spend over 100 million dollars on autism research. The 2005 WSJ article noted the controversy over Simons' funding:
"When the Massachusetts Institute of Technology asked him for money for brain research, he demanded that the project focus on autism and include scientists he liked. He has provided his family's DNA for study, pitched in to help solve research problems and is pushing scientists to probe a genetically based explanation for the disease."
"Many are cheering this influx of cash, hoping Mr. Simons' riches can buy a breakthrough. Others complain that Mr. Simons isn't working with existing autism groups and that his focus on finding a genetic explanation could miss the disease's true cause."
Yesterday, two years later, the WSJ article cautions:
"by suggesting that something within parents' control could be triggering autism, Prof. Waldman has reopened old wounds in the realm of autism research, which is littered with debunked theories linking the disorder to the family environment."
The WSJ quoted senior vice president of Autism Speaks and mother of an autistic child, who said: "Autism is a genetic disorder. The only thing the parents do wrong is they have bad genes." Autism Speaks was founded by two years ago by Bob Wright, Vice Chairman and Executive Officer and GE Chairman of NBC Universal, whose grandson is autistic. However, if Waldman's research was solid, why shouldn't economists study autism? Wouldn't parents appreciate a solution in hand, as opposed to one that entails decades of research and development?
The connection between autism and TV would best be studied in controlled experiments between groups of children, but according to the WSJ, economists don't have the "money or the access to children" to perform this kind of research. The WSJ quoted "Ami Klin, director of the autism program at the Yale Child Study Center, [who] says Prof. Waldman needlessly wounded families by advertising an unpublished paper that lacks support from clinical studies of actual children." In other words, Klin said, Waldman's conclusions conflicted with results of clinical trials that were already done.
The genetic links to autism are currently being studied intensely and the effort is fruitful. Scientists have discovered genes that could account for one or more aspects of Autism Spectral Disorder. The Yale Child Study Center is partially funded by the Simon's Foundation.
The WSJ also quoted Klin sayng: "The moment you start to use economics to study the cause of autism, I think you've crossed a boundary." Yet is the question really about whether economists can study science problems? Economists contribute significantly to fields including psychology, ecology, and international development.
Scientists distort the issue by focusing on parental blame, or whether an economist can contribute to research. Shouldn't we just look at whether a specific paper more approximates rigorous research or Swiss cheese? Waldman's paper was criticized because it drew speculative conclusions and was advertised in a what amounted to a sensationalist press release as opposed to being published in a peer reviewed academic journal. While the researchers and authors may have been swayed by conviction, their resulting study didn't meet the standards of the autism community, psychologists or neurobiologists. As WSJ reported, Joseph Piven, director of Neurodevelopmental Disorders Research Center at the University of North Carolina, said of the confounding variables, "It is just too much of a stretch to tie this to television-watching...[W]hy not tie it to carrying umbrellas?"
Did the paper meet economist's standards, a skeptical reader's standards, or for that matter the own researcher's standards? Scientists ideally start from a neutral position then work to disprove their theories, in order to prove them to themselves, their peers, and the world. Social scientists approach problem solving similarly. If the paper was representative of the field of economics, we might look at economics and its influence with renewed skepticism.
Why not just conclude that this particular exploration, however well intended, wasn't that rigorous, and/or didn't seem to support the author's conclusions and final recommendations. Waldman perhaps used his reputation in another field to build media interest around this hypothesis. He's not the first researcher to flip the scientific process on its ear. Other scientists have announced "results" prior to publication, with various motives. Perhaps his move was strategic, but it remains to be seen whether this economist can circumvent the research process to successfully demand that the science community study his hypothesis.
-----------------------------------------
¹ The original paper "Does Television Cause Autism?", is posted online at NBER working papers and from the Cornell website. Read it for yourself. What do you think?
²Last October, Acronym Required wrote "Autism, TV, Precipitation: Dismal Science", a satirical 10 step research how-to for repeating the results of Waldman's original paper.
Leave a comment