It is not race or class that makes communities more susceptible to industrial pollution. The reason that environmental justice research has produced “very mixed results,” says Don Grant, a sociologist at the University of Arizona, is that it’s been asking the wrong questions.

People, from sociologists to activists to policymakers, “like to reduce problems like pollution to a single factor, such as race or income. But our findings suggest it’s more complex than that,” and should include traits of the polluting firm, says Grant. “It doesn’t make sense to focus on one particular variable; it makes more sense to talk about these things coalescing in certain ways.”

Grant and his colleagues used data from the U.S. Environmental Protection Agency’s Risk-Screening Environmental Indicators on the toxic emissions of individual facilities and their associated health dangers. They appended to these data not only neighborhood characteristics such as race and income, but organizational features like the size of the facility. Then the researchers employed a novel statistical technique called “fuzzy set analysis,” which, instead of simply assuming linear and straightforward causes, allows for the unexpected.

“What we’re finding is that there are multiple pathways to the same dangerous outcome,” says Grant. By changing the framework, Grant and colleagues resolved many of the field’s paradoxes. It’s not that one study is wrong and another is right, he says. “Our study is showing that poverty and minority presence do have inconsistent effects: They are important in some contexts and not in others.”

Each of the recipes for risk that Grant identified validates seemingly contradictory, prior case study findings. For example, one of the most potentially hazardous combinations is a large absentee-owned plant in an African-American neighborhood. Another is a neighborhood with both large African-American and Latino populations—a contributing factor other studies have described as ethnic churning. A third is the interaction between being poor and being African American.

“It’s not race or class, it’s both, and it can be both in different ways in different kinds of neighborhoods and in relation to different kinds of firms,” says Scott Frickel, an environmental sociologist at Washington State University who reviewed Grant’s paper. Frickel thinks the new method “actually gets us a lot closer to what’s really going on out there.”

This is more than just theoretically important. “When you can identify the types of facilities that are dangerous in certain types of communities, that lends itself to some kind of policy remedy,” says Grant. It could help regulators to focus their efforts on the most potentially harmful facilities, firms to focus their improvements on the most efficient pathways, and scholars to focus their research on the most fruitful case studies. “There’s an entirely different way to look at this,” says Frickel.

Don Grant, Mary Nell Trautner, Liam Downey, et al., “Bringing the Polluters Back In: Environmental Inequality and the Organization of Chemical Production,” American Sociological Review, 75, 2010.

Read more stories by Jessica Ruvinsky.