Purple loosestrife, zebra mussels, house sparrows, Asian carp—these are all non-native invasive species decried widely among those who seek to support native ecosystems and their flora and fauna. But does such partisanship belong in science?
“Context bias in invasion biology,” by Robert J. Warren II, Joshua R. King, Charlene Tarsa, Brian Haas, and Jeremy Henderson, appeared recently in PLOS ONE. Warren, associate professor of biology at SUNY Buffalo State, said, “We wanted to see if the use of value-laden language indicates that researchers are biased in ways that influence their results.”
The article weeded 650 peer-reviewed studies about invasive species competition from more than 1,300 studies and coded data with the intent of determining if scientists’ biases against invasive species were reflected in pejorative language and ultimately in how they interpreted their data.
“We expected to find that boilerplate language with value-laden terms suggesting that invasive species are ‘bad’ or ‘harmful’ has increased in published papers,” said Warren. However, they found that such language peaked in the mid-2000s, reaching 70 percent in 2006–2007, but has decreased steadily since then.
“That’s encouraging,” said Warren. “It shows that science is self-correcting.” The study suggested that the decrease in biased language may have been affected by a number of highly critical papers from scientists within and outside of invasion biology in which “researchers argued for less outright disdain and language vilifying invasive species.” Warren said that the implication is that science may not passively correct itself: instead, scientists must practice self-criticism and avoid defensiveness.
Another hypothesis examined was whether observational studies, in which correlational data are interpreted by the researchers, were more like to report that invasive species outcompete natives than experimental studies, in which the results are less subject to interpretation. Observational studies were indeed more likely to report invasive species as better competitors than native species, suggesting that the interpretation of results may have been affected by the researcher's bias.
"The bad news is that we found that bias apparently influenced results in scientific papers," Warren said. "The good news is that we found the problem self-correcting, likely because of rigorous scientific debate and self-criticism. The hallmark of science is relentless self-criticism."