Halo effect

What it is

The halo effect is the tendency for one overall impression of a person, brand or product to shape how we judge its separate qualities. If we like someone, we tend to rate them higher on traits we have little evidence about. The reverse also happens: one bad impression can drag everything else down, sometimes called the “horns” effect.

An everyday example

Two neighbors offer to look after your plants while you are on vacation. One is warm, chatty and always says hello in the elevator; the other is quiet and a bit abrupt. Without thinking much about it, most people would trust the friendly neighbor to be more reliable, more careful with watering and even more knowledgeable about plants.

None of those qualities has been observed. Friendliness is real information about friendliness, but it has been stretched to cover traits it says little about.

Why it happens

Forming a single overall impression is quicker than evaluating each quality independently. Once that impression exists, it acts as a lens: specific traits are filled in to match it. Importantly, people are often not aware this is happening and believe their specific judgments are independent of their overall liking (Nisbett & Wilson, 1977).

How to counter it

• Rate one thing at a time. When assessing people, such as in interviews, score each criterion separately and, if possible, before forming an overall verdict.
• Ask what evidence you actually have for each trait, rather than what fits your general feeling.
• Separate the messenger from the message. Judge an idea or product on its own merits, apart from who presents it or how attractive the packaging is.

What the research shows

The term comes from Thorndike (1920), who noticed that supervisors’ ratings of the same employees or officers on quite different traits, such as intelligence, physique and leadership, were much more strongly correlated than seemed plausible. He described this as a “constant error” in ratings.

In a later experiment, students watched one of two videos of the same instructor, who spoke with an accent. In one he behaved warmly, in the other coldly. Those who saw the warm version rated his appearance, mannerisms and accent as appealing; those who saw the cold version rated the same features as irritating. Participants denied that their liking had influenced these ratings (Nisbett & Wilson, 1977).

A practical caveat: some traits really are correlated, so not every overlap in ratings is a halo. This makes the effect hard to measure precisely, even though the experimental evidence that it exists is clear.