Opinion: Are Americans really dumb enough to worry about food containing DNA?

In case anyone wondered, all of the food in this French market--and everywhere else--contains DNA.
(Marlene Awaad / Bloomberg)

A recent university survey found that 82% of Americans support the mandatory labeling of genetically engineered food. No big surprise there; a perennial argument by advocates for labeling is that most people want things that way. (Labeling initiatives were rejected by voters in California and Oregon.)

But here’s another interesting finding of the poll, one of a monthly series of so-called Food Demands Surveys by Oklahoma State University: Almost as many Americans, 80%, support “mandatory labels on food containing DNA”—which all food has because it all comes from living things with genetic material.

The information has been making the rounds of social media as a fun way to mock the American public for its ignorance of science and its largely unfounded fears that bioengineered food is unsafe to eat. Those dummies, the story goes, don’t even know what DNA is.

But let’s hold the laughter and groans for a moment. Repeated past surveys have found that most Americans know quite well what DNA is and that genetic material is, in a way, the very nature of food. It’s unlikely that everybody was suddenly stricken with stupidity in 2015.


As blogger Ben Lillie notes, the question about DNA came in a long list of questions that were almost solely about government policies regarding the safety and wholesomeness of food. So people who favor protective policies were in a way primed or lulled into saying yes unless they looked and questioned closely. Further, because people are accustomed to the notion that GMO food involves DNA that has been tinkered with in a laboratory, it’s very likely that they read the concept of lab-engineered into that question. In other words, whether the writers of the poll intended it that way or not, it’s a trick question.

Two other phenomena that can affect poll results: Respondents tend to want to please the pollster by answering in ways that they perceive are sought, even though they might not be aware of it. And, of course, almost everything is free on a poll. It’s easy to be in favor of pastured farm animals during a poll; the question is whether people will still support it when they find out that the resulting meat will cost three times as much.

Another aspect of the survey that calls for some skepticism is the fact that it is an online poll, not a random telephone survey. Most online polls lack the kinds of crucial safeguards that ensure the results represent the public as a whole; the materials published by the university don’t indicate what the procedures and protocols were, except to say that the results were weighted to reflect the demographics of the American public. The pollsters were probably off for Martin Luther King Jr. Day and couldn’t be reached by phone.

The bigger question than demographics, though, is whether the pollsters had a way of reaching a truly random audience. If the poll involves people logging on of their own accord, that means it’s more likely to attract people who care deeply about the subject and far more likely to draw activists. Word can pass quickly through social media. Another question: Were there safeguards to keep the same people from logging on over and over again, under different identities?


The real lesson here might be less about American ignorance of science—though there are many other signs that people don’t understand many crucial aspects of the field—but polling, and how often it might not indicate what we think it does. There are, of course, great polling operations that take all possible measures to reduce the margin of error to the minimum, but any really surprising poll results should first prompt a more careful look at the poll itself.

The results might, in their own way, tell us something about why attempts to require labeling and growth of GMO foods fail more often than they succeed. Advocates often give the poll results as a reason to pass labeling requirements. Polls show that the public wants it, they say. And when the measures fail, they blame the agriculture industry for spending so heavily against the measure. There is no doubt about it: The food companies shell out loads of money on aggressive counter-campaigns. But if support for restrictions on GMO food were really that strong to start with, would these measures lose anyway?

Perhaps. But it’s also possible that the polls primed respondents to answer in a certain way; that people were unconsciously guessing what the poll sought and responding appropriately; and that they didn’t think about any of the counter-arguments until there was an actual vote.

Follow the Opinion section on Twitter @latimesopinion