STATS ARTICLES 2009
STATS responds to queries from the Milwaukee Journal Sentinel
Trevor Butterworth, July 7, 2009
Given the extent of STATS criticism of the Milwaukee Journal Sentinel’s “Chemical Fallout” series on the chemical bisphenol A (BPA), and that the authors of the series have now contacted STATS to inquire about our motivation and funding for taking on this issue, we have decided that transparency and dialogue are best served by posting our responses to the paper’s inquiries. Below is the full text of the responses sent to both reporters.
Dear Ms. Kissinger, Ms. Rust:
My responses to the questions you submitted are as follows:
1) According to IRS 990 forms, stats.org received $100,000 in donations in 2007. That same year, the Sarah Scaife Foundation donated $100,000 to stats.org. Is stats.org’s funding solely from the Sarah Scaife Foundation?
My only function at STATS is to edit and write for stats.org. I am not involved in funding activities or decisions. Therefore, in response to your query, I am passing along the following information from STATS management:
STATS is not solely funded by the Sarah Scaife Foundation. They were the only donor who contributed in FY2007, providing general support of $100,000. During FY2008, however, STATS received $100,000 from the Stuart Family Foundation, $40,000 from Mr. Paul Mongerson, and a $70,000 contract from the Endocrine Society for work related to a planned conference on the Women's Health Initiative.
2) How long were you working on this report?
I spent three to four months writing, reporting and researching “Science Suppressed.” But I have been following and reading on the BPA issue for the last few years, because it illustrates problems with the media’s coverage of scientific data more generally
3) Did you receive funding from any other source while working on this story? In other words, were you contracted to do this? Or did the Scaife funds provide the monetary support you needed to complete your report?
I received no funding from any other source other than STATS, for which I work on a contract basis (that is, I am not a salaried employee). I decided myself to write the piece, given the power vested in me as editor of the site. No one suggested or directed that I take on this project. If anything, there was concern that this was eating up too much of my time relative to my other duties as editor and frequent contributor to stats.org. Incidentally, since the organization had several funding sources during the time I was actually working on this story in 2009, it would be incorrect to say that the Scaife funds provided the monetary support.
4) You relied heavily on two figures in your report: Robert Chapin and Calvin Willhite. Were either of these scientists paid by your organization or any other organization or person to contribute to your story? Also, why did you choose Dr. Willhite and Dr. Chapin? Are you concerned that these scientists might be characterized as having a bias?
First, no source received any direct or indirect payment of any kind from STATS for contributing to the report.
Second, I do not think that any reading of the piece supports the contention that I relied heavily on Chapin. He has a couple of quotes in a 27,000-word piece – and they have more to do with his interaction with the Milwaukee Journal Sentinel than the nature of the science. Willhite, Dekant, and EFSA itself provided more in the way of scientific analysis.
Third, STATS is all about statistical method: these scientists’ work reflected the weight of evidence as determined by the statistical methods noted by the National Toxicology Program in 2001, as well as other experimental methodologies agreed on by international bodies and protocols. We don’t attribute bias based on guesswork or inference, we look at method.
Willhite, an expert on reproductive toxicity, conducted a full overview of the research for NSF International – which is about as independent as you can get. He was slightly more cautious than EFSA on a reference dose, but the fact that they concurred on so many points merely underscored that there were good, objective criteria for disentangling good from not so good research.
Thus, when Chapin notes that you can’t ignore or dismiss the route of exposure, he is reiterating the broad consensus of assessing the risk of BPA.
5) You often refer to stats.org is an independent, non-partisan group. How do you reconcile this description when funding for your organization comes from a very partisan source?
Being independent and nonpartisan means following the evidence rather than catering to any party, ideology, or interest group. STATS does receive part of its funding from a prominent conservative foundation. But the notion that the data follow the dollars is something that needs to be proven rather than assumed. Anyone expecting STATS to follow the conservative “party line” might be surprised at the number of pieces that challenge widely-held conservative contentions.
For example, STATS contributors have documented the consensus of climate scientists on anthropogenic global warming, supported the findings of controversial Lancet studies which attributed a high civilian mortality rate to the American invasion and occupation of Iraq, challenged the costs and efficacy of the “war on drugs,” and uncovered excesses by “boot camps” practicing “tough love” for troubled teenagers (and advised bipartisan congressional hearings which led to regulation of the industry). And they have written for a wide range of publications, including progressive outlets such as the Huffington Post, Salon, and Mother Jones.
Certainly STATS’ criticism of the media for overplaying chemical health risks comports with the views of many conservatives (and non-conservatives); however, across the full span of STATS’ work there may be less “ideological” consistency than is displayed by some of the environmental organizations on whose work the Milwaukee Journal Sentinel relied in its BPA reporting.
In addition, STATS’ statistical analyses are vetted by our research director, Dr. Rebecca Goldin, prior to publication. Dr. Goldin is a Harvard and MIT-educated mathematician who currently teaches at George Mason University. She is a member of the Science Policy Committee of the American Mathematical Society and a recipient of the Ruth Micklir Prize, which is awarded annually to an outstanding female mathematician. Dr. Goldin and I had lengthy discussions about the BPA issue and whether there were serious scientific and statistical grounds for the claims being made.
6) We have noted that you frequently comment on blogs, stories, and other contributions on the web that are on the topic of bpa. How long has this chemical been an interest of yours? Why?
STATS’ mission is to investigate how science and statistics are used and abused in the media and public policy. I became interested in BPA largely because journalists have frequently ignored the weight of evidence in research on this chemical, along with the global regulatory analysis, in favor of a small group of scientists whose claims have been stronger than their statistical rigor.
My interest was piqued by the 2006 publication of the European Food Safety Authority’s risk assessment of BPA. It received virtually no media attention, despite recommending a five-fold increase in the daily tolerable intake. The European Union has been highly cautious about chemical risks, employing the precautionary principle, so this presented a disconnect with the U.S. media coverage, which seemed worth examining.
The genesis of “Science Suppressed” can be traced to a call we received in January 2008 from a reporter with ABC7/KGO-TV News in San Francisco. She was looking for an independent perspective on BPA; I suggested that she find a toxicologist and focus on the competing experimental methodologies. She spent weeks reading the key documents and interviewing sources, including STATS Research Director Dr. Rebecca Goldin. The final (Emmy-nominated) story included an interview with NSF Board Member Dr. Calvin Willhite, who spent two years working on calculating a reference dose for BPA, which meant he surveyed all the key evidence for and against the chemical.
When I read Dr. Willhite’s (peer-reviewed journal) findings, I noted that he reached pretty much the same conclusion as the EU’s risk assessment had. Yet the U.S. press coverage continued to highlight studies claiming that BPA was a serious health risk. So I began to think about writing an in-depth piece on this topic. In May 2008 I discussed writing a story on BPA for Gourmet magazine. But a combination of other commitments and stream of new studies on BPA derailed the piece.
When I finally had time to think seriously about it in January 2009, I felt I needed to do a much more comprehensive piece first, one that would ask the experts missing from the media narrative what they thought – and one that subjected journalistic accounts to expert scrutiny. It was at this point that I began seriously researching the story. At the same time, STATS was engaged in a survey of members of the Society of Toxicology about chemical risks, and it made sense to wait and see what they thought before publishing anything.
The survey, which was released at about the same time as my story, found that only nine percent of toxicologists rated BPA as a high risk to health, compared to 26 percent who rated sunlight as a high risk and 29 percent who saw a high health risk in aflatoxin, a naturally occurring fungus found in peanut butter. So we had empirical data to show that alarmist media coverage of the chemical was at odds with the views held by most scientific experts.
In addition, toxicologists were brutally critical of the media coverage of chemicals. Large majorities rated a wide range of media outlets (as well as environmental groups) as overstating the risks of chemicals. When 87 percent of toxicologists say that news coverage of chemical risk is not balanced, 90 percent say that journalists fail to seek out diverse sources in writing about chemical risks, and 97 percent say the media do a poor job of distinguishing good studies from bad studies, there is something terribly wrong. Findings like these also suggest that my interest in critiquing media reports on BPA and other chemicals is also in the public interest.