Personality test questions: what they look like and how to answer

What does a personality test question look like?

On a trait-based test, most of the questions are not questions. They are short statements about how you usually behave, and your job is to say how accurately each one describes you. The commonest response scale has five points of accuracy: "Very inaccurate", "Moderately inaccurate", "Neither accurate nor inaccurate", "Moderately accurate" and "Very accurate". Some inventories use agreement instead, from strongly disagree to strongly agree. The wording differs; the task does not.

The statements are deliberately dull, and that is care rather than laziness. A good item names one behaviour, in the present tense, in words a tired person can answer in a few seconds.

• One idea per sentence. A statement saying you are organised and punctual cannot be answered honestly by someone who is one and not the other.
• Ordinary words. An item that needs a second reading measures reading comprehension along with personality.
• Typical behaviour, not rare events. Items ask what you usually do, because one unusual evening is a poor guide to a lifetime of them.
• No verdict built in. Items describe rather than praise, so agreeing is not a boast and disagreeing is not a confession.

What are some example personality test questions?

Here are eight real statements from the item bank behind the free Snapshot described at the end of this page. Each belongs to a broad dimension, and each is scored in a direction the person answering never sees.

Read alone, each statement is almost banal, and none is trying to catch you out. It is the averaging of several, and the comparison of your average with other people's, that turns ordinary sentences into a score worth reading.

Why are some questions reversed, and why so many of them?

Reversed items exist because of a habit called acquiescence: some people agree with most statements put to them, whatever the statements say. If every item ran in the same direction, that habit would masquerade as the trait. Mixing directions cancels most of it, because someone who genuinely agrees with "Am the life of the party" should disagree with "Don't talk a lot". Agreeing with both says more about how the form was filled in than about the person.

The number of statements per trait is the other piece of quiet engineering. One item is mostly noise: it carries whatever the respondent took a particular word to mean, plus their mood, plus the phrasing. Reliability rises as you add items tapping the same underlying trait, which is the reason serious inventories are longer than they look as though they need to be (Nunnally & Bernstein, 1994).

The cost of going too short has been measured rather than guessed. Across samples of 437 employees and 355 students, very abbreviated Big Five measures led researchers to understate how much traits matter for behaviour, single-item measures worst of all, while slightly longer scales improved validity substantially at very little cost in time (Credé et al., 2012).

Where do personality test questions come from?

Many of the statements on free Big Five tests trace back to the International Personality Item Pool, a public-domain collection built so the broad dimensions could be measured without licensing a commercial inventory. Seven measurement researchers set out the case for public-domain scales, with that pool as their prototype, in the paper usually cited for it (Goldberg et al., 2006).

Public domain means what it sounds like: the items can be read, copied, criticised and reused by anyone, which is why an almost identical statement turns up on several sites. Commercial inventories keep theirs under copyright, which is ordinary practice, though it does mean you cannot inspect the questions before answering them. Being free to copy is not the same as being well made: what matters is which items a test kept, how many per trait, and whether it says what they measure. How the Snapshot was built and scored works through those decisions.

How should you answer personality test questions?

The instructions that produce the most useful score are short. You are describing a habit, not making a case.

• Answer as your typical self over the last year. Not your best week, not your worst, and not the version of you that turns up for a first meeting.
• Go with the first reading. The answer you would give in three seconds usually describes the habit better than the one you reason your way to in thirty.
• There are no correct answers. A high score is a position, not an achievement, and a low score is the other end of an ordinary human dimension.
• Do not answer as the person you want to be. It is the commonest way to make a report useless, and it flatters on every dimension at once.
• Use the middle option when it is true. Plenty of statements genuinely do not fit either way, and saying so is information. Using it to avoid deciding is not.
• Answer in one sitting, with nobody reading over your shoulder. Both of those move answers more than people expect.

The questions will feel repetitive, and they are meant to. Several statements circle the same trait from slightly different angles so the average survives one odd answer (Credé et al., 2012).

What if the test is for a job?

Answer it honestly. A profile assembled from the answers you think an employer wants describes an imaginary person, and if it works you have won a job that suits someone else. Deliberate self-presentation is also not subtle: it shows up as an unusually flattering pattern across every dimension at once.

It is also worth knowing what a hiring test can do. Trait scores predict work outcomes on average and across many people rather than one at a time, and no careful publisher claims otherwise. Wellington is not designed for hiring or selection. Career personality tests covers what the work-related versions measure.

What do these questions miss?

Every statement above asks you about yourself, and you are a good but uneven judge. In a study of 165 people rated by friends, by strangers and against behavioural criteria, the self was the best judge of internal traits that are hard for anyone else to see, friends were the better judges of evaluative traits such as intellect, and for visible traits such as extraversion self and others did about equally well (Vazire, 2010).

That is a reason to hold some parts of a report more loosely than others, not to distrust the whole thing. Pooling 263 samples covering 44,178 people, observer ratings predicted behaviour about as well as self-ratings, and better for some outcomes such as academic achievement and job performance (Connelly & Ones, 2010). If a result surprises you, show the trait page to someone who has known you for years.

How does Wellington ask its questions?

Wellington is a science-based personality assessment from Therabot Labs LLC. It measures personality as continuous traits, built on the Big Five and HEXACO models, and writes the results back as a personal report rather than a type.

The free Snapshot asks 60 statements, 10 for each of the 6 dimensions it reports (Extraversion, Agreeableness, Conscientiousness, Emotional Stability, Openness to Experience, Honesty-Humility), each rated from "Very inaccurate" to "Very accurate", and takes about 7 minutes. Reversed statements are flipped before scoring, each dimension is the average of its own items, and that average is reported as a percentile against the reference sample, in one of three bands: Quiet below the 30th percentile, Balanced to the 70th, High above it. The bands and the Portrait names exist to make the result easier to talk about; the percentile is the measurement, and it is always shown alongside. Our norms are provisional while the norming sample is completed, and the report says so.

The first ten questions are answered before anything is asked of you, so you can read the statements and decide whether the style suits you. After question ten you sign in with an email address, so your answers are saved and your results can be shown to you. There is no card. The membership adds a further 336 statements to reach 90 traits, including every Big Five facet and all 24 character strengths. Wellington is a wellness tool for reflection, not a medical or psychological diagnosis, and it does not replace care from a qualified professional. It is not designed for hiring, and you can export or delete your answers at any time.

Questions people ask

Are there right answers on a personality test?

No. The statements describe ordinary behaviour, and both ends of every dimension are positions rather than results. Nobody passes or fails. The only way to answer badly is to answer as someone else, because the score then describes a person who does not exist, and the report written from it will not fit you.

Should I answer as I am at work or at home?

Answer as you are across an ordinary year, taking both settings together. Most people behave somewhat differently at work, and a trait score is meant to capture the average rather than one setting. If the two feel far apart on a dimension, that gap is worth noticing, and it usually shows up as a middle score.

Why do the questions feel repetitive?

Because each trait is measured several times over. A single statement carries the respondent's mood and their reading of one particular word, so scores built on one item per trait are noisy. Reliability rises as items measuring the same trait are added (Nunnally & Bernstein, 1994), which is why a careful test covers the same territory from several angles.

How many questions should a personality test have?

Enough for several statements per trait. The authors of a widely used ten-item measure said plainly that it is inferior to a standard inventory and should be chosen only when time is severely limited (Gosling et al., 2003), and very short measures understate how much traits matter (Credé et al., 2012). Around ten per broad dimension is a common compromise.