We all like to think we're rational and swayed by science, but there's often one big problem with this commitment to evidence-based living -- namely, that the science we're working with is way out of date, or just plain wrong.

Think about it: the last time most non-specialists boned up on the facts was in high school or college, and a lot has happened since then. That's why many people missed the news that spinach was never really very high in iron (apparently, a simple decimal point error is to blame for this myth) or that vitamins are essentially a boondoggle.

That might also be why you insist on having unstructured interviews with job candidates. It can't be because the practice is actually useful, after all. As Yale management professor Jason Dana pointed out recently in The New York Times, pretty much all the available evidence shows that informal interviews are completely useless.

Science says interviews are a waste of time.

How can Dana be so sure? There's no shortage of scientific evidence to back him up. Experts have had serious doubts about the effectiveness of interviews for decades.

Way back in 1979, a natural experiment saw the University of Texas Medical School at Houston admit 50 additional candidates who had previously been eliminated during the interview phase of the application process, for instance. The students admitted after flunking the interview performed exactly as well as the students who had passed it.

"The judgment of the interviewers, in other words, added nothing of relevance to the admissions process," writes Dana.

Dana's own research backs up these earlier findings. A series of studies he conducted with colleagues revealed that students actually did worse at predicting the GPA of fellow students after they interviewed them, as opposed to just looking at their past grades and current courses. The student interviewers even failed to notice when interviewees gave entirely random answers based on the first letter of the last words of each question.

Why won't people give up on informal interviews?

The fact that the interviewers failed to notice the person they were chatting with was responding totally randomly is worrying, but there was an even more alarming section to the study. The students making predictions also said they still valued interviews even after the researchers had explained that those interviews were sometimes random and were making the students' predictions worse.

Why do students -- and many, many managers -- insist on ignoring evidence and doing the workplace equivalent of downing hundreds of dollars in useless vitamins? People love a good story, and interviews provide a more interesting and memorable narrative about candidates than a dry resume. The only problem is these narratives are often wrong.

If hiring managers were rational, they'd quit their useless unstructured interview habit for good. If that's beyond their courage, Dana suggests they "structure interviews so that all candidates receive the same questions, a procedure that has been shown to make interviews more reliable and modestly more predictive of job success" or "use interviews to test job-related skills, rather than idly chatting or asking personal questions."

Failing those two options, Dana pleads with hiring managers to at least show a little humility and take it to heart that science proves interviews are pretty much hocus-pocus.

Are you convinced that most interviews are utterly useless?