RV_TaleOfTwoDataSets_1One of the biggest obstacles we faced in publishing our groundbreaking report, “Vermont’s GMO Legacy,”  was the collection of reliable data. It was no small feat, even with the advantage of having the Vermont Agency of Agriculture, Food, and Markets (AAF&M) requiring farmers and pesticide and fertilizer applicators to report their usage.

The long, nearly year-and-a-half road from first obtaining the data, analyzing it, and publishing it more often felt like we were writing a novel than a research report, all the result of the many problems we uncovered from the state’s handling of the raw data. It was so bad, in fact, that the state even re-issued the data at one point after admitting that they had grossly mishandled it.

We dedicated an entire section of the report to our “methods of analysis,” in which we attempted to address some of the state’s data drama. Here’s our recap:

Unfortunately, the Vermont data set, which was analyzed and shared with several legislators, the Secretary of Agriculture, and staff from Ben & Jerry’s was found to have been corrupted by faulty data entry and a poorly designed computer program. According to officials at the AAF&M, the data entry staff miscalculated the quantity of chemicals, and the computer programmers were only counting the amount of the primary pesticide in mixed products that have multiple pesticides as active ingredients since 2008. Since a great majority of the pesticides used in Vermont are mixtures of two or more chemicals, this oversight was very significant. The data entry errors overestimated the recent use of some pesticides, but more importantly, the computer programming errors significantly underestimated the total tonnage of pesticides applied on Vermont corn.

In a commendable effort to correct these errors, a second set of data was provided by the AAF&M and analyzed for this report. All of the pesticide data analyzed in this paper is derived from the second set of data. 

Once we ironed out the problems and traded data sets back and forth with the Vermont Agency of Agriculture, we still communicated with state officials and independent researchers for months, all in an effort to make very, very sure that the data we were using was absolutely correct. And then, just to be extra certain, we even sent a near-final draft of our report – data, analysis and conclusions – to state officials, just to make sure, once again, that the information they provided us was correct. They said it was.

But one major problem still exists after we published the paper on June 28th: The AAF&M has not updated its own website, thus still reporting the original, inaccurate figures on pesticide use. We have requested repeatedly that they either update or remove this inaccurate data, especially as our report has gone public and will draw significant attention to it.

For those familiar with the State of Vermont’s many snafu’s with regard to its other web-related endeavors (Vermont Health Connect, for example), you won’t be surprised by the answer we received: They can’t update the data because they don’t control the website. Go figure.

As a result, it didn’t take long for someone who doesn’t agree with our report’s conclusions (get off GMOs!) to do a comparison between the correct data in our report and the incorrect data still published by the state. Gotcha! they screamed, incorrectly.

It is, to say the least, a rather awkward position we find ourselves in, publishing the correct data but still vulnerable to the inaccurate accusations made by those who are using the state’s still-published inaccurate data. Oddly enough, we’re in the same boat as the state in our inability to control the data on its website.

We will, however, continue to pressure the state to figure out a way to get control of – and correct — the data it is presenting to the public. In the meantime, we will do what we can to get the word out about this great tale of two data sets.

And, just to make sure that everyone who reads and analyzes our report understands that the state officials who oversee the data are not questioning our numbers, we offer this passage from VTDigger’s exclusive coverage of the report’s release:

“Cary Giguere, chief of the Agriculture Resource Management Division, [said] the data used in Allen’s report is correct…”