How long does a personality test take, and how long should it?

How long do the common personality tests take?

The table below gives questions and an approximate reading of minutes. Treat the minutes as rough: reading speed, how carefully you weigh each statement and whether you are on a phone all move the figure, and most published instruments report item counts rather than a stopwatch time. A useful rule of thumb is four to eight short self-report statements a minute once you have settled into the rhythm.

Why does length trade against reliability?

A personality questionnaire estimates something it cannot see directly. Each statement is a noisy indicator: you might read one quickly, or think of an unusual week, or interpret a word differently from the next person. Averaging several statements cancels much of that noise out. This is the oldest result in psychometrics, and it has a formula: the reliability of a scale rises as items are added, quickly at first and then with diminishing returns (Nunnally & Bernstein, 1994).

The practical consequences of ignoring it are well documented. When very short Big Five measures were compared directly with full-length ones, the short versions substantially understated how much traits influence important behaviours and overstated the importance of newer constructs, raising the risk of error in both directions (Credé et al., 2012). The authors of one widely used ten-item measure said much the same thing when they published it. It is adequate when questionnaire space is severely limited, and a longer instrument should be preferred whenever there is room for one (Gosling et al., 2003).

There is a second cost that is easy to miss. Length is what buys you resolution, not just precision. Five domains can be measured in sixty questions. Thirty facets underneath those domains cannot, because each facet needs its own items. If a test reports fifteen or thirty specific traits from forty questions, it is splitting the same evidence more finely rather than gathering more of it.

What is the shortest test that still works?

For the five or six broad dimensions, the working minimum is somewhere around eight to ten items per dimension, which is where the well-regarded free inventories sit. The BFI-2 uses 60 items for five domains (Soto & John, 2017). That length is not a coincidence: it is about where reliability stops improving fast enough to justify more questions.

• Two to four items per trait: fine for a research study with no room, not enough for a report about you.
• Eight to twelve items per trait: the standard for a broad-dimension inventory, and enough for a stable band.
• Twelve or more per trait: what facet-level instruments use when they want to report thirty specific traits rather than five general ones.

If a test does not say how many questions it asks per trait, divide the total by the number of traits it reports. That one division tells you more about what the result is worth than any claim on the landing page.

What if you only have five minutes?

Then take a short test, read the result as a rough orientation, and plan to take a longer one later. A five-minute result can tell you which dimensions you are clearly high or low on. It cannot reliably tell you the difference between the 45th and the 60th percentile, and it should not be used to make a decision about a job, a relationship or a course of study.

Three ways to spend a short window well. Answer for your typical self over the past year, not for today, because a single hard week tilts self-ratings on Emotional Stability in particular. Answer at a steady pace and resist going back to rewrite, because the items ask about your typical behaviour rather than about a particular occasion, and a long deliberation tends to be a search for exceptions. And do not take a short test twice in a row expecting the second run to settle it, because a scale with two statements per trait gives the averaging almost nothing to work with.

Can you pause a personality test and come back?

It depends entirely on whether the test saves your answers, which in practice means whether you are signed in. A test that runs in the browser without an account usually keeps your progress only in that browser tab, so a closed tab or a flat battery costs you the sitting. A test with an account can save each answer as you give it and hand the sitting back to you on another device.

The general advice is to finish a single scale in one sitting where you can, because your frame of reference drifts between days. The person answering on a calm Sunday is slightly more generous about their own patience than the person answering on a hard Wednesday. Splitting a long instrument by chapter is a reasonable compromise, since each chapter still covers its traits in one consistent mood.

How long does Wellington take?

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 is 60 questions and takes about 7 minutes: ten statements for each of the 6 dimensions, which is the eight to twelve items per dimension a broad inventory needs. The first ten questions come before anything is asked of you. After question ten you sign in with an email address, so your answers are saved and your results can be shown to you. From that point everything autosaves, so you can close the tab and come back on your phone. There is no card. At the end every trait 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, with a written page on each dimension. 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. Those norms are provisional while the norming sample is completed. 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 data at any time.

The Portrait, which the Wellington Membership adds, is 336 questions in 4 chapters, about 36 minutes in total, and it is designed to be taken a chapter at a time over several days. It brings the report to 90 traits. The membership then asks for about four minutes a week, one short module at a time, working toward all 255 traits across 104 weeks, with retests of the earliest traits so you can see what has moved. Retakes are always fresh sittings; you cannot edit old answers, and your previous sittings are kept so you can compare.

Questions people ask

Is a 10 question personality test accurate?

It is accurate enough to tell you which dimensions you are clearly high or low on, and no more. With two items per trait there is very little averaging to cancel out a misread statement, and relationships with real outcomes come out weaker than a full inventory would show (Credé et al., 2012). Treat a ten-item result as a signpost.

How long should I spend on each question?

A few seconds. Self-report personality items are about your typical behaviour, and your first reading is usually your most honest one. If a statement genuinely splits in two for you, because it is true at work and false at home, answer for the setting where you spend more of your life, and note the split for yourself.

Can I stop halfway and finish later?

In Wellington, yes, once you are signed in. Your answers autosave from that point, and the Portrait's 4 chapters are meant to be taken on different days. Before you sign in, your answers live only in your browser, so finish the first ten questions in one go. Other tests vary, and many keep no progress at all.

Why is the Wellington Snapshot 60 questions rather than 10?

Because 60 questions is ten per dimension, which is roughly where reliability stops improving quickly, and because each band comes with a written page that has to be worth reading. Ten questions across 6 dimensions would be under two items each, which is enough for a label and not enough for a report.