Personality tests compared: what each one actually measures

With personality tests compared side by side, the main difference is what each one tries to capture: broad traits, types, workplace styles, strengths, interests or skills. The Big Five has a large, well-tested research base as a general description of personality. Other tools can be useful too, as long as you know what they were built to do and what the evidence does and doesn't show.

No single test covers all of you. Picking one is more like picking the right map for a trip: a road map and a hiking map can both be good, for different journeys.

Traits vs types: the difference that matters most

Before getting to names, it helps to know about one big split.

• Trait measures place you on a scale. You might be somewhat above average in one trait and well below in another.
• Type measures sort you into a category. You are one type or another.

Types are easy to remember and fun to talk about. Scales hold more information, since most human differences seem to come in degrees. Two people on either side of a type boundary may be almost identical, yet get different labels.

The Big Five

What it measures: five broad traits: openness, conscientiousness, extraversion, agreeableness and emotional stability.

How it reports: a score on each scale.

The evidence: Lewis Goldberg's 1992 paper developed markers for the five-factor structure, and those markers are the basis of the public-domain IPIP items we use. The Big Five also holds up when tested against real life. Christopher Soto attempted to replicate 78 links between these traits and life outcomes, and 87% succeeded in the expected direction, with effects typically 77% as strong as the originals.

Good for: a general picture of your personality. Less good for: a catchy label. There is no neat type to put in your bio.

The Myers-Briggs Type Indicator

What it measures: preferences on four pairs of opposites, which combine into one of 16 four-letter types. It draws on Carl Jung's theory of psychological types.

How it reports: a four-letter type.

The evidence: Robert McCrae and Paul Costa studied it alongside the five-factor model in a sample of 468 adults. They found no support for truly dichotomous preferences or distinct types, and they found that the instrument measures four relatively independent dimensions. Those four indices did relate to four of the five major personality dimensions. David Pittenger reviewed its psychometric limits and urged considerable caution in drawing inferences from the four-letter formula.

Good for: starting a conversation about differences, which is how many people use it. Less good for: treating your type as a fixed fact about you.

DiSC

What it measures: behavioural styles, built around four letters: Dominance, Influence, Steadiness and Conscientiousness. Its publisher, Wiley, describes the model as resting on two dimensions, and it is often used in workplaces.

How it reports: a style or a blend of styles.

The evidence: we have not reviewed the research on this one in depth, so we don't make a claim about it here. If it is being used for a decision at your work, ask the provider for their technical evidence.

Good for: talking about communication at work.

The Enneagram

What it measures: nine personality types.

How it reports: a primary type, usually with some additional detail.

The evidence: Joshua Hook and colleagues published a systematic review of the research, covering 104 samples, and they describe the evidence for reliability and validity as mixed. Factor-analytic studies usually found fewer than nine factors, and there was little research support for secondary elements of the system, such as wings. Some studies suggested people found it helpful for personal or spiritual growth.

Good for: reflection, if you find that useful. Less good for: making measured comparisons between people.

CliftonStrengths

What it measures: Gallup, its publisher, describes 34 themes grouped into four domains: strategic thinking, relationship building, influencing and executing.

How it reports: your themes, with your strongest ones first.

The evidence: as with DiSC, we have not reviewed this research in depth and so we don't make a claim here. It is a commercial tool designed around what you do well, not a general map of personality.

Good for: talking about strengths. Less good for: seeing the whole of you, because that isn't what it sets out to show.

RIASEC career interests

What it measures: John Holland's six types of vocational interest: Realistic, Investigative, Artistic, Social, Enterprising and Conventional.

The evidence: Margaret Nauta's review reports considerable empirical support for the RIASEC types and how they are ordered. Support for the idea that a good match between you and your job predicts outcomes is present, but its predictive power appears modest.

Good for: generating career ideas. Our Career Compass Expedition is built on this theory, using Selfcarta's own questions, which are not yet independently validated.

Emotional intelligence

What it measures: it depends on who you ask. John Mayer, Peter Salovey and David Caruso describe a schism in the field between emotional intelligence as a set of mental abilities and as a broader mix of positive traits. So two tests with the same name can be measuring different things.

Good for: reflection on how you handle emotions. Check which kind of test you are taking.

How to choose

• Want a general picture of your personality? Start with a Big Five measure.
• Want to talk with your team? A style tool can open a useful conversation, as long as it isn't used to judge anyone.
• Thinking about careers? Try an interests measure.
• Want to know how you come across? No self-report test can fully tell you that. You need other people's input, which is what Second Sight is for.

If you are not sure how far to trust any online tool, read are online personality tests accurate? and our notes on the science behind Selfcarta.

Start with the Big Five

If you only take one, the Big Five is a sensible first map. The Core Personality Expedition uses public-domain IPIP items based on the Big-Five Factor Markers, and it is free to take with no sign-up.