According to conventional wisdom, simply getting a foot in the door is a step toward equality for women and people of color in fields dominated by white men. If hiring managers actually consider and interview women and non-whites, then women and non-whites have a good chance of actually getting ahead on their merits, right?

Maybe not. A series of studies described in a recent Harvard Business Review article indicate that having a single woman or a single person of color in the finalist pool for a job is effectively equivalent to having zero women or people of color. “If there’s only one woman in your candidate pool, there’s statistically no chance she’ll be hired,” write business professors Stefanie K. Johnson and David R. Hekman and Ph.D. candidate Elsa T. Chan.

Johnson, Hekman, and Chan suspected that since “people have a bias in favor of preserving the status quo,” they’d be more likely to select candidates who conform to the status quo—which, in most business settings, means white men. So they asked undergraduates to evaluate the job applications of three equally qualified candidates and select one. But the researchers manipulated the “status quo” by altering the race and gender makeup of the fake applicants: Some groups of applicants comprised two white men and one woman or black man; the others comprised two women or two black men and only one white man. (The researchers signaled race and gender by changing the names of the fake job applicants, varying between the black-sounding “Dion Smith” and the hilariously white-sounding “Connor Van Wagoner,” among other made-up names.)