The halo effect
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.
In 1920, an army researcher asked commanding officers to rate their soldiers on separate qualities: intelligence, physique, leadership, character. Logically these should vary independently, a strong soldier need not be a smart one. But the ratings did not come out that way. If an officer saw a soldier as physically impressive, he tended to rate that same soldier as more intelligent, more capable, more honest, across the board. One strong impression bled into every other judgement. Later researchers found the effect runs below awareness: in one study, students who saw a lecturer behave warmly rated his appearance and accent more favourably, and denied that his warmth had influenced them. One good trait lifts how you rate all the others.
A single strong quality in a person, their confidence, their looks, one impressive credential, lifts your judgement of everything else about them, and you will not feel it happening. This is why the candidate who dazzles in conversation gets rated as more competent, more reliable, more everything, on no evidence beyond the dazzle. The defence is to judge one thing at a time and write down each judgement before you form an overall impression. Score the actual work separately from the person's presence, because the presence is trying to colour the work, and it is very good at it. The most common source of that presence is charisma, which is learnable.
The halo effect says one strong quality in a person lifts your rating of everything else about them, on no evidence beyond that quality. The Pygmalion effect says what you expect of a person changes how you treat them, and how you treat them changes what they become. They start from the same first impression and end in different places. Halo corrupts the score while the person stays the same. Pygmalion moves the person while the score stays honest. A manager who believes in someone gets both at once and cannot tell which one they are watching. The test is whether the outcome moved or only the rating did, and the only way to know is to have written the rating down before the outcome arrived.
Source: Thorndike, A Constant Error in Psychological Ratings, Journal of Applied Psychology, 1920. Confirmed: Nisbett and Wilson, 1977, Journal of Personality and Social Psychology.
Phil Rosenzweig, 2007
How one strong impression shapes the rest of the story, in business especially.
The Saturday letters
Every Saturday, one idea worth your time, with an honest account of how far the evidence really goes and where it stops. Free, a few minutes to read, and it leaves you with questions rather than instructions.
Draw your own card. It does not take long, and it rewards taking your time.