Good Science Bad Science Part I
May 3, 2012
A lot of scientific studies are done in the field of nutrition, often with conflicting or misleading results. It’s important for people to know the difference between a well-designed study and one that’s not. It’s also important that people be able to determine the validity of conclusions drawn from a study. Most important of all, don’t trust others to interpret a study—read it for yourself.
To illustrate, I’ll take two studies—both done at Harvard. Let’s start with the colony collapse disorder study overseen by biologist Chensheng Lu. He starts with the question: Why are honeybee colonies suddenly collapsing? He proceeds to review the scientific literature on the subject. That’s good science. Then comes the hypothesis: Colony collapse disorder is caused by honeybee exposure to imidacloprid, an insecticide used on corn and transferred to bees in high fructose corn syrup fed to them by beekeepers. Next step: Design a study to test the hypothesis.
In this case new honeybee colonies populated by healthy bees were established in central Massachusetts. Some were fed HFCS containing various concentrations of imidacloprid, and some received HFCS that was free of the pesticide. All the bees were treated the same except for one variable—the pesticide dose. None of the field workers knew the concentrations of pesticides the bees were getting. Good science. The experiment is described in the final report in such detail that anyone who doubts the results could replicate it. Good science.
The results supported the hypothesis. The greater the pesticide dosage, the faster a colony collapsed, and all of the pesticide-exposed colonies eventually failed. Three out of four of the control hives (no pesticide) survived. So the data collected strongly support the hypothesis. That’s what scientists hope for but don’t always get. Next, the authors suggest questions raised and avenues of study for other scientists researching the same topic. Good Science. Finally, the study is published in a reputable scientific journal where other scientists can learn from it and possibly challenge it. Good science
Here we have the elements of a good scientific study whose results we can take seriously. In the next post we’ll take a look at a bad nutrition study done at the same university.



Ken Olsen of Ground Works talks about laying the groundwork for a school garden so that people work as a team and keep the garden going in the long term. One Sioux Falls school garden he describes has been going for four years.
Here’s one of my basic life principles: Every technological ‘advance’ brings with it one or more negative effects.