Sources & methodology

This report puts your 24 strengths in an order. That order comes entirely from your own answers, measured against your own average — not against anyone else’s.

The reason is simple: there is no public dataset of scores for these 24 scales. Plenty of research uses them, but no large set of real responses is published anywhere we could use, so there is nothing to work out percentiles from. Rather than invent cut points and present them as if they meant something, the report says only what the answers can support — which of these 24 sounded most like you, relative to the rest of your own answers.

So “first” on your list does not mean you are more curious, or kinder, than the people around you. It means curiosity or kindness came out ahead of your own other 23 answers.

Each strength gets a score: the average of your answers to its questions, on the 1 to 5 scale. Before the strengths are put in order, your own average across all 24 is subtracted from each one.

That step does two useful things. It stops a generally warm answerer and a generally cautious one from getting different results for the same underlying pattern — if you answered high to almost everything, what matters is which ones you answered highest. And it keeps the number attached to a claim we can make: how this strength compares with the rest of you.

It does not change the order. Subtracting the same figure from all 24 moves every score by the same amount, so the sequence stays exactly as your answers put it.

The report names five leading strengths, and sometimes six or seven. That is not a mistake — it happens when your scores are exactly level at the boundary.

If the strength in fifth place is tied with the ones just behind it, all of them are shown, up to 7. Cutting the list at five would mean breaking a genuine tie on an arbitrary rule and then presenting the result as though it were a finding. If more strengths are level than would fit, the report keeps the group at 5 and names the others underneath, so nothing is hidden.

Ties are ordinary here. With 24 scales scored on averages of 4 questions each, two strengths landing on the same number is common rather than remarkable.

Worth naming, because it looks like a contradiction. The instruction before the questions is the standard one for this item pool, with one change: it asks you to describe yourself in relation to people you know around your age. The original also asks you to picture people of the same sex as you, and we took that half out, because we never ask your sex and nothing in this report is worked out separately by sex.

The published reliability figures below were obtained under the original wording, so that is a real change from the conditions those figures were gathered under, and worth knowing when you read them. The instruction only shapes how you answer a single question. What the report then does with your answersis a separate step, and that step compares your 24 answerswith each other — never with another person’s.

96 questions across 24 strengths, from the International Personality Item Pool — a public-domain collection of personality items maintained by Lewis Goldberg at the Oregon Research Institute. The 24 strengths themselves are the Values in Action classification described by Christopher Peterson and Martin Seligman.

The particular 4 questions used for each strength are a short set chosen from that pool by Matthias Bluemke and colleagues in 2021, who picked them to cover each strength evenly, to work across countries, and to run two questions each way. There is a longer version, 213 questions rather than 96, and we used it until September 2026. We moved because the longer one asked about some strengths almost entirely one way round and others almost entirely the other, which let a habit of agreeing move one strength up your list past another. The trade for that is real and is in the reliability figures below: fewer questions per strength means each strength is measured less precisely.

• 96 questions, 4 per strength.
• 48 of them are worded the other way round — exactly two on every strength, against two worded straight. Agreeing with a reversed one counts against the strength rather than for it.
• That balance is the reason for this version of the questions. A reader who leans toward agreeing with everything gains as much from the two straight questions as they lose on the two reversed ones, on every strength equally, so the habit cannot push one strength up the list past another.
• No two questions in a row come from the same strength.
• Item wording is unchanged from the source, apart from a closing period added to one question, which is recorded in the item bank.

You may have seen these strengths sorted into six virtues — wisdom, courage, humanity, justice, temperance and transcendence. This report doesn’t use them, for two reasons.

The item pool itself publishes no grouping at all: its own table lists the 24 scales alphabetically, with no categories above them. And the six virtues were, in the words of the people who publish them, arrived at by reasoning rather than found in the data. When researchers have gone looking for groups in real answers, they have generally found three to five, not six, and not the six.

Showing them anyway would mean giving a structure more authority than it has earned. So the 24 are presented flat.

Reliability here means internal consistency — whether the 4 questions within one strength tend to move together. The published figures run from 0.57 to 0.81. Each strength was measured twice, on a German sample and a British one, and the figure we quote is always the weaker of the two. Appreciation of Beauty, Humility, Judgment, Kindness, Prudence, Self-Regulation and Teamwork sit below the 0.70 mark often treated as a floor — a mark set with much longer scales in mind, but worth knowing. Self-Regulation is the weakest of the 24 at 0.57, and a placing for it carries less weight than the rest.

The figure above is omega rather than the more familiar alpha, and that is the source’s own choice rather than ours. Alpha assumes things about a scale that are not true of these — a five-point answer scale, and questions balanced two each way — so the people who published these scales print it beside them marked as misleading.

The same people asked a few hundred of their respondents to answer again two to three weeks later. Agreement between the two sittings ran from 0.53 to 0.77, again taking the weaker of the two samples. That is the figure to hold in mind if you take this twice and your order comes out differently: some movement between sittings is expected, and the strengths at the top of your list move least.

One caveat on all of it. Both samples were European, one German and one British. Nobody has published figures for these scales on an American sample, so the numbers above describe how the questions behaved there rather than here.

Reliability matters more on this report than on a report of positions, and it is worth being plain about why. When a score is a little off on a banded report, someone reads a slightly wrong description. When a score is a little off here, a strength moves up or down the list — so two strengths a place or two apart are not meaningfully different, and the difference between your fifth and your sixth is often smaller than the measurement error. Read the top of your list as a group, not as a sequence.

• It does not say you lack the strengths at the bottom. All 24 are strengths you have; the list is their order in you.
• It does not say how you compare with anyone else, and nothing in it can be read that way.
• It does not predict what you will do, or how well you will do it. These are self-descriptions, not tests of ability.
• It is not fixed. Answers move with circumstances, and some of these strengths — energy in particular — move more than others.