Assess oral work

Body language in presentations: assessing it without caricature

Published September 11, 2026 · Lire en français

In short — "93% of communication is non-verbal" — you've probably heard this claim. It's widely misquoted and has nothing to do with assessing student presentations. That doesn't mean body language is irrelevant: a few observable criteria (grounding, eye contact, speech rate) genuinely matter and can be taught. Here's how to assess them honestly, without penalizing students who simply don't fit a dominant cultural style.

The 93% myth: what Mehrabian actually said

In the 1960s, psychologist Albert Mehrabian ran studies on how people judge whether someone likes them, based on facial expressions and tone of voice. He found that in that narrow context, body language and tone outweighed the words themselves. What he never said is that 93% of all human communication is non-verbal.

Applied to a student presenting a history project, the figure is meaningless. Yet it keeps appearing in teacher training sessions, coaching workshops, and even some official rubrics. The result: students are penalized for "bad posture" while their actual argumentation — the point of the exercise — barely counts.

Start by being honest about what non-verbal actually contributes to a classroom presentation: it's real, but modest compared to content and structure.

Three criteria that are genuinely observable and teachable

Rather than trying to assess "charisma" or "presence" (which are subjective and often culturally loaded), focus on the three non-verbal elements students can actually work on:

• Grounding: standing still with a stable base, not swaying or shifting weight constantly. Restlessness signals anxiety to the audience; stillness signals confidence. This is trainable with simple exercises.
• Audience contact: looking at the audience rather than reading off notes or staring at the board. Not necessarily making prolonged eye contact with individuals — distributing gaze across the room is enough, and it's far less culturally loaded than insisting on direct one-to-one eye contact.
• Speech rate: not rushing. Most students accelerate when nervous, swallowing key points. A deliberate pace gives listeners time to follow — and it sounds more confident, because it is.

💡 These three criteria have one thing in common: they're observable from the back of a room. If you can't note it while standing in the doorway, it probably doesn't belong on a grading rubric.

What not to assess: the cultural trap

Several non-verbal behaviors that get treated as "professional standards" are actually culturally specific and should not appear as graded criteria without careful framing:

• Direct eye contact: highly valued in North American and North European contexts, often perceived as aggressive or disrespectful in others. Assessing "looks the audience in the eye" without nuance can systematically disadvantage students from different backgrounds.
• Gestures: some students gesture a lot, others barely at all. Neither is wrong. Unless gestures actively distract (self-touching, pen-clicking), leave them off the rubric.
• Smiling and general "likeability": completely subjective and often gender-biased. Drop it.

The goal is to assess the presentation, not the person. This matters especially for shy students or those with anxiety — a topic worth exploring in our article on helping nervous students at oral exams.

How to make non-verbal criteria explicit and fair

Whatever criteria you choose, three conditions make them fair:

1. Announce them in advance. Students should be able to prepare. "During your presentation, I'll be looking at whether you maintain a stable stance and whether you make audience contact" is fair. Surprising them is not.
2. Keep the weight modest. Non-verbal criteria should represent a small portion of the overall grade — perhaps 10–15%. The rest should go to content, structure, and communication of ideas. Inverting this sends the wrong message about what counts.
3. Distinguish criterion from criterion. Don't bundle everything under "presence" or "oral delivery." One row on the rubric for audience contact, one for speech rate: separable, observable, gradeable. For guidance on building this kind of rubric, see our article on how to create an oral presentation rubric.

Using practice rounds to work on non-verbal

Non-verbal behavior changes slowly — and almost never from a single comment. What actually moves things is repeated low-stakes practice: short rounds where students focus on one target criterion per pass. "This time, focus only on your speech rate" is actionable. "Work on your presence" is not.

Peer observation is particularly effective here. Assign one student to watch only the speaker's feet (grounding), another to count how many times the speaker looks away from the audience. The act of observing one criterion sharpens the observer's awareness of it — they start self-monitoring the same thing in their own presentations. This is the core of peer assessment for oral presentations.

💡 Record a two-minute clip of each student (with their consent) during practice rounds. Watching yourself is uncomfortable but fast — most students immediately spot what they couldn't feel while speaking.

Respecting individual styles

Every good speaker has a personal style. Some are still and precise; others are energetic and expressive. The goal isn't to produce a single model of "good presenter." It's to ensure students can communicate their ideas clearly enough that the audience can follow.

A student who gestures broadly and moves across the room isn't worse than one who stands perfectly still — they're different. As long as the gestures don't distract and the speech rate is manageable, both presentations can be fully effective. Your feedback should improve their specific style, not push everyone toward the same mold.

This is also why grading bias matters here: the same non-verbal behavior can read very differently depending on a student's gender, origin, or even whether the evaluator is tired. Knowing the well-documented patterns — halo effect, contrast effects — helps you calibrate. Our article on grading bias in oral assessment covers the main traps.

A practical checklist for your rubric

• ✔ Are each non-verbal criterion observable from a distance?
• ✔ Did students know these criteria before the presentation?
• ✔ Does non-verbal represent less than 20% of the total grade?
• ✔ Have I checked each criterion for cultural assumptions?
• ✔ Can I describe the difference between levels without using words like "natural," "confident," or "engaging"?

Wrap-up

Body language in presentations is real, but modest — and far more useful as a teaching target than as a grading criterion. Focus on three observable behaviors (grounding, audience contact, speech rate), announce them upfront, and keep them proportional. Drop what's culturally loaded or subjective. And give students repeated short practice rounds — that's what actually moves non-verbal behavior, far more than any single detailed comment.

Frequently asked questions

Is the 93% non-verbal rule real?

No. Albert Mehrabian's figures apply only to the expression of feelings and attitudes in specific conditions — not to communication in general. Applying them to student presentations is a misquote that leads to absurd rubrics.

Which body language criteria are actually teachable?

Three that are observable and can genuinely improve: grounding (stable base, not swaying), eye contact distributed across the audience, and speech rate (not rushing). These are concrete enough to target in practice.

How do I avoid cultural bias in body language assessment?

Separate what you can observe (pauses, speech rate, audience contact) from what you're interpreting culturally (direct eye contact, gestures). Announce your criteria beforehand so students can prepare, and always let content and structure carry more weight than non-verbal style.

Should body language be a graded criterion?

It can be, but only if it's genuinely observable and communicated in advance. A light criterion worth a small percentage is enough — this signals it matters without penalizing students whose personal style is simply different from the dominant model.