Confirmation bias in evaluation
Researchers have tested this many times, in different places, over many years, and they keep finding the same result. This one is safe to trust.
Researchers told people that a stranger they were about to interview was either an extrovert or an introvert, assigned completely at random. Then they let each person choose which questions to ask. The people who believed the stranger was extroverted picked questions like, what would you do to liven up a party? The ones who believed the stranger was introverted asked things like, what makes it hard for you to open up? Neither group set out to be unfair. But the questions they chose were built to confirm what they already believed, and so, naturally, the answers came back confirming it. The extrovert questions produced extrovert-sounding answers. The interviewers had, without noticing, manufactured the evidence for a belief that had been assigned by a coin flip.
When you have a theory about someone, you unconsciously ask the questions that will confirm it, and then you treat the confirmation as proof. You read a CV, form an impression, and walk into the interview quietly testing that impression rather than the person. The way out is to deliberately ask what would prove you wrong. Decide in advance what a strong candidate who contradicts your first impression would say, and go looking for that. Otherwise the interview is a mirror of your opening guess rather than a test of the person.
Confirmation bias says that once you have a theory about someone, you ask the questions that will confirm it, and the answers you manufacture feel like evidence. The Pygmalion effect says that once you have an expectation of someone, you give them more or less to work with, and the result you produce feels like a prediction coming true. Both begin with a private guess about a person. One rigs the evidence. The other rigs the outcome. Confirmation bias ends in an interview that was a mirror of your opening impression. Pygmalion ends in a team member who failed partly because you had already written them off. In both cases the guess was never neutral, and the fix is the same: decide in advance what would prove you wrong, and go looking for it.
Source: Snyder and Swann, 1978, Journal of Personality and Social Psychology.
Daniel Kahneman, 2011
On how the mind seeks agreement with what it already believes, and calls the result evidence.
Draw your own card. It does not take long, and it rewards taking your time.