No, it's not true that only 7% of your meaning is communicated by your words - Mehrabian Miscommunication

I've heard this so many times, I completely assumed it was true - and passed it on to other people. It goes that when you're communicating, 55% of the communication comes from your body language, 38% from your tone of voice, and only 7% comes from the actual words you use. It sounds really compelling and makes a great case for paying attention to your delivery and posture. 

Where did that rule come from?

The original paper was by Albert Mehrabian in 1967. He and his team were interested in whether the way you say words affects how people interpret your meaning. They had three groups;

1. They asked 25 people to look at a list of 15 'neutral' words like 'maybe' and imagine someone saying them to another person. They were asked to rate on a scale of 1-9 how much they thought the speaker liked the person they were speaking to. 
2. They asked 17 people to listen to recordings of someone saying those words in a positive, neutral and negative way, and to rate how much they thought the person speaking liked the person they were speaking to in each case. They then asked those people to look at photos of the speakers saying the words in the three ways and rate how much they thought the speaker liked the person they were speaking to.
3. They asked 20 people to listen to the recordings and look at the photos at the same time in a random order, so that the negative/neutral/positive tone didn't match up. Then they asked them to rate how much the speaker liked the person they were speaking to. 

The researchers took the people's ratings and did some statistical analysis, combining these results with another study. They found that the people decided the 'level of liking' most strongly from the speaker's facial expression, then from their tone, and lastly from the word they said. They calculated the 'weight' of each part to be 55%, 38% and 7% respectively. 

Does this mean that communication is split 55-38-7%?

Does that mean that when you're trying to communicate, 55% of your meaning comes from your facial expression? Well, no. And I think most people instinctively feel like that doesn't match up with their experience, which is why the Mehrabian miscommunication is so interesting to people. There are a few reasons why the rule isn't backed up by this study, including;

• All of the people in the study were women, all undergrads of the same university who participated for credit on a psychology course. 
• There was only 62 of them. 
• The study had people communicate single words with no context. 
• The study didn't even look at body language, just facial expression.
• The study wasn't looking at whether the speaker successfully communicated the word - they were asking about whether or not the speaker liked the person they were speaking to. 

What useful conclusions does the paper actually have?

There are a couple of conclusions that you could take from this study. If your face, tone and words match up, you communicate more effectively - the ratings were more consistent when everything had the same emotional quality. So if you want people to get your message in presentations, pay attention to whether or not your face or your tone of voice is supporting what you're trying to say. And if your face, tone and words don't match up, people will assume your facial expression is telling the truth. If you've ever had a friend tell you they're fine with a scowl on their face, you know this one already. 

Why is the Mehrabian miscommunication so pervasive?

This sort of thing happens all the time across scientific research. Oversimplification of studies by the media leads the general public (like me) to believe all sorts of truisms which aren't true at all, and aren't even what the scientist was trying to say. Having lots of numbers attached to the rule makes it sound official and statistical, and easier to remember. And the basic premise - that you should pay some attention to both content and delivery - is true. 

In the end...

It isn't the end of the world whether or not the numbers actually back up the premise. This rule is used to demonstrate that our communication isn't just down to the words we say, which is hard to argue with. What matters is that we're not misusing science to support our conclusions, or worse, inadvertently giving reason for people to distrust science and academia.