There is a widespread myth in the area of public speaking and presentations, that “we only get 7% of information from words, while the majority comes from tone of voice (38%) and body language, facial expressions/gestures (55%)”. This myth has been falsely based on the research of Albert Mehrabian, Professor Emeritus of Psychology at the University of California, Los Angeles.
Albert Mehrabian is clearly tired of how people keep twisting his study to suit their own narratives so he wrote on his website:
If a person does not talk about their feelings or relationships (i.e., like-dislike), these equations [7%-38%-55%] are not applicable
— Albert Mehrabian
So, if anyone ever tells you again that “we only get 7% of information from words, while the majority comes from tone of voice (38%) and body language, facial expressions/gestures (55%),” shoo that “expert” away. This is a myth based on the fact that people simply haven’t bothered to read through all of Mehrabian’s research carefully.
In two of Mehrabian’s experiments, they tested how people give an emotional evaluation to a message when the communication is inconsistent—meaning, you might see something positive in the person’s face, but the audio comes across as negative (tone of voice), while the actual words are neutral. Simple, right.
For instance, in the 2nd experiment, participants were shown a photo of someone looking hostile, while they played a recording of the word “maybe” in different tones (neutral, positive, negative). Then they’d ask: “Do you like this person or not?” Obviously, the answer is going to be no, because he looks hostile.
In the 1st experiment, they’d play a recording of just one word—let’s say, “Thank you” in an aggressive tone, or “Terrible” in a positive or neutral tone—and then ask: “What do you feel in this message?” Of course, in that context, intonation will matter more. It’s just like in Russian when someone says “Well, that’s just fantastic”, it could be genuine praise or condemnation, depending on the tone—even though, by the words alone, it sounds like praise. In English, we have “You are killing it, dude!” or “That is dope as hell!”—the words might look one way, but the meaning really hinges on how you say them.
Again, Mehrabian’s study covers only inconsistent communication and only emotional evaluations by the listener. If someone’s looking for a phone with a processor frequency above 2 GHz, and in a video review somebody neutrally/hostilely/positively says, “The iPhone processor is 3.23 GHz,” then those “7%-38%-55%” equations don’t apply. Which basically means most business communications don’t fall under these ratios at all.
You can check out Mehrabian’s original study for details: https://psycnet.apa.org/doiLanding?doi=10.1037%2Fh0024532
Links to other debunkings of this myth:
1. The 7-38-55 Rule: Debunking the “Golden Ratio” of Conversation
2. European Polygraph 2019 No2 (AMSEL)