Self-other agreement by Big Five dimension: where the gaps are biggest

When researchers compare how people rate their own personality to how others rate them, they find agreement — but far from perfect agreement. More interestingly, the degree of agreement varies substantially across the five major personality dimensions. Extraversion is relatively easy to agree on. Neuroticism is not. Understanding why these gaps exist, and what they mean in practice, is essential for anyone using personality data for development.

This article covers the self-other agreement literature in detail, dimension by dimension. For the broader case for why peer ratings are needed at all, see Why self-assessment alone isn't enough. For how Cèrcol's Witness instrument operationalises peer assessment in practice, see what the Cèrcol Witness instrument measures.

What Self-Other Agreement Research Actually Measures

Self-other agreement studies work by collecting personality ratings from two sources: the person themselves (self-report), and one or more people who know them well (informant ratings — in Cèrcol's framework, Witnesses). The two sets of ratings are then correlated. A correlation of 1.0 would mean perfect agreement; 0 would mean no systematic relationship at all.

In practice, the correlations fall in the range of .25–.65 depending on the dimension and the relationship between rater and ratee. The landmark meta-analysis by John and Robins (1993) — drawing on data from hundreds of participants — established the basic pattern that has been replicated many times since. Their data are grounded in the Big Five tradition, specifically in the NEO-PI framework, and their findings have held up across cultures, languages, and assessment methods.

For background on the broader field of self-knowledge and its limits, see: Self-knowledge (psychology).

"The magnitude of self-other agreement varied significantly across the five factors, with Extraversion showing the highest agreement and Neuroticism the lowest, consistent with the observability hypothesis."
— John & Robins (1993), Journal of Personality and Social Psychology, doi:10.1037/0022-3514.63.1.146

The Observability Hypothesis: Why Some Traits Show Bigger Gaps

The central theoretical explanation for differential self-other agreement is the observability hypothesis: dimensions that involve visible, external behaviour should show higher agreement than dimensions that involve internal states, motivations, or experiences that are not directly observable.

This is intuitive. If you want to know whether someone is sociable, you can watch them at a party. If you want to know whether they experience chronic worry, you cannot observe that directly — you can only observe downstream behavioural signals that may or may not appear.

The observability hypothesis predicts the pattern we actually find in the data: Extraversion (high observability) at the top, Neuroticism (low observability) at the bottom. The intermediate dimensions fall in between, with their placement explained by how visible the relevant behaviours are.

Self-Other Agreement Dimension by Dimension: Full Breakdown

Presence (Extraversion): highest agreement

Extraversion — Cèrcol's Presence — consistently shows the highest self-other agreement across studies, with correlations typically in the range of .55–.65. The reason is that extraversion is fundamentally about behaviour in social situations: talking, initiating, leading, energising. These behaviours are directly observable by anyone who interacts with the person.

If you rate yourself as highly extraverted, your Witnesses can verify or disconfirm this by observing your behaviour in meetings, social events, and collaborative work. The alignment between internal experience and external expression is high for this dimension.

This also means that disagreements on Presence are particularly meaningful. If you rate yourself as highly present but your Witnesses rate you as low, that divergence is worth examining. It may indicate a gap between intention and behaviour — you feel engaged and energised internally, but it is not coming across.

Discipline (Conscientiousness): moderately high agreement

Conscientiousness — Cèrcol's Discipline — shows moderate to good self-other agreement, typically .45–.55. Discipline is partly observable: work outputs, punctuality, organisation, follow-through are all behaviours that Witnesses can observe and assess. But the internal aspect — the effort, the self-regulation, the attention to detail in private work — is not visible.

A self-rated high-Discipline person may feel that their internal standards are rigorous, even when external outputs don't fully reflect this (due to resource constraints, context, or life circumstances). Witnesses rate what they see: delivered work, reliability, organisation of shared spaces. These two perspectives both capture real aspects of conscientiousness, but they are not identical.

Bond (Agreeableness): intermediate agreement

Agreeableness — Cèrcol's Bond — is interesting precisely because it sits at the intersection of the internal and the external. How warm, trusting, and cooperative you are involves both internal dispositions and visible behaviour. The visible aspects — whether you help, cooperate, and avoid conflict — should produce agreement. The internal aspects — your actual feelings of warmth and trust — are private.

Self-other correlations for Agreeableness typically fall around .40–.55. The complexity here is social desirability: Agreeableness is highly valued in most cultures, and both self-ratings and observer ratings may be elevated by the desire to appear (or to perceive others as) kind and cooperative. This complicates the interpretation of gaps.

A low Bond rating from Witnesses may reflect genuinely low agreeableness — or it may reflect that the person's warmth is not being expressed in a way Witnesses can read. Context matters enormously: someone who is warm and collaborative in close relationships may come across as cool and guarded with acquaintances.

Vision (Openness): intermediate agreement

Openness to experience — Cèrcol's Vision — shows correlations in the range of .40–.50. This intermediate agreement reflects a genuine complexity in the dimension itself. Openness captures both observable behaviours (creative output, intellectual curiosity expressed in conversation, novel pursuits) and internal states (richness of inner experience, aesthetic sensitivity, unconventional thinking).

Someone who is highly open internally — who has a rich imagination, who connects disparate ideas privately — may not fully express this in contexts where creative output is not invited or valued. Their Witnesses will rate what they observe, which may underestimate the depth of their Vision. Conversely, someone who performs openness (eclectic taste, conspicuous intellectual interests) may rate themselves as more open than their actual cognitive style warrants.

Depth (Neuroticism): lowest agreement

Neuroticism — Cèrcol's Depth — consistently shows the lowest self-other agreement, with correlations often in the range of .25–.40. This is the most important finding for practical personality assessment, and it warrants careful attention.

Neuroticism involves the tendency to experience negative emotions — anxiety, sadness, irritability, vulnerability. These are predominantly internal experiences. A person with high Depth may be experiencing considerable emotional turbulence that is simply not visible to Witnesses, particularly in professional contexts where emotional display norms are strong.

There is also an asymmetry: people who experience high negative affect often work hard to conceal it, which is a form of emotional regulation that is itself part of the high-Depth profile. The result is that high-Depth individuals often appear calmer to Witnesses than they feel internally — producing low self-other agreement not because either perspective is wrong, but because they are measuring different things: internal experience versus external expression.

The practical implication is significant. Depth (Neuroticism) is the dimension that shows the greatest gap between how people experience themselves and how they are experienced by others. This is precisely where self-report data is least sufficient, and where peer ratings — even imperfect ones — add the most value.

Agreement Levels by Big Five Dimension: Summary Table

What Self-Other Gaps Mean for Team Development and Coaching

In team settings, these differential gaps have direct practical implications.

For Presence, the high self-other agreement means that a person's self-rating is a fairly reliable guide to how others experience their energy and social engagement. Divergences here are meaningful signals.

For Depth, the low agreement means that you should be cautious about assuming you know how much emotional load someone is carrying. The colleague who appears calm and collected may be managing considerable internal distress. Teams that create space for honest discussion of emotional experience — rather than relying on observed behaviour — are more likely to surface this.

For Bond and Vision, the intermediate agreement is a reminder that both self-ratings and Witness ratings capture partial truths. A person who feels highly collaborative (high self-rated Bond) but is rated as less agreeable by Witnesses may be experiencing a disconnect between their intentions and their perceived behaviour. That gap is a coaching opportunity.

For a structured approach to turning these gaps into a development plan, see using Cèrcol for team development: a practical guide. And to understand how each dimension connects to specific team roles, see the 12 Cèrcol team roles explained.

How Forced-Choice Methods Reduce the Self-Perception Gap

One factor that inflates apparent agreement in traditional Likert-scale assessments is shared response biases. Both self-raters and peer raters may systematically use the upper end of the scale (acquiescence) or may both be influenced by the evaluative valence of traits (social desirability). This shared bias artificially inflates self-other correlations — the agreement you observe is partly genuine, but partly an artefact of both parties using the scale in similar ways.

Forced-choice methodology, as used in Cèrcol's Witness instrument, removes acquiescence bias by construction: you cannot agree with both options. It also reduces social desirability effects by pairing adjectives of similar valence. This means that forced-choice data captures genuine personality differentiation more cleanly — and that the self-other agreement figures derived from forced-choice assessments are more accurate reflections of true perceptual alignment.

For a full discussion of the forced-choice approach and its psychometric advantages, see Blind spots in teams. For the question of how many Witnesses you need before the composite is reliable, see how many peer assessors do you need for reliable personality data?

Summary: Which Big Five Gaps Matter Most and Why

Self-other agreement is highest for Presence (Extraversion) and lowest for Depth (Neuroticism), with Discipline, Bond, and Vision falling in between. The primary driver of this pattern is observability: behaviours that are visible produce more agreement; internal states that are not directly observable produce less. Understanding which dimension shows the largest gap in a given individual is itself valuable information — and it shapes how Cèrcol interprets and presents peer personality data.

References
John, O. P., & Robins, R. W. (1993). Determinants of interjudge agreement on personality traits: The Big Five domains, observability, evaluativeness, and the unique perspective of the self. Journal of Personality and Social Psychology, 63(1), 146–156. doi:10.1037/0022-3514.63.1.146