Sitting-rising test
Sit down on the floor from standing, then get back up, leaning on as little as you can. You start at ten and pay for what you lean on. The rubric is the published one, the arithmetic is done here, and the score is out of ten.
This is the one instrument on the site that is not Limber’s. The app does not run this test and does not score it; the scale below belongs to the researchers who published it, and the reading it gives you stays outside the Range Score.
How to take the reading
- Clear a space on a floor that will not slide. Barefoot, in clothes you can move in, with nothing to trip on behind you.
- From standing, sit down cross-legged on the floor, using as little support as you can manage. Take it slowly. Stop at a stretch, never at sharp pain.
- From that position, stand back up, again with as little support as you can.
- Count the supports you used going down and the supports you used coming up, separately, and note whether you wobbled in either phase.
A support is anything you put weight through that is not your feet: a hand, a forearm, a knee, the side of a leg, or a hand pressed on your own knee. Leaning on furniture or a wall counts too. Each support costs one point in the phase it happened in, and a visible loss of balance in a phase costs half a point.
The test was published as a supervised assessment, taken with somebody watching and space to fail safely. Read it that way: if getting to the floor is something you are unsure about, have someone with you, and if it is something you would rather not risk at all, do not take it. Not taking a test is a legitimate answer, and it is a better one than a fall.
No reading yet. Both phases have to be answered before there is a score: half a test is not a five, and this page will not fill in the half you did not take. Choose the supports you used going down and coming up, add the wobble if there was one, and press Keep this reading.
Sitting-rising test scoring, in full
Two phases, five points each, so the test starts at ten and every deduction is something that actually happened. One point comes off for each support used in the phase it was used in; half a point comes off for a visible loss of balance in that phase. The form above does the subtraction and prints the working.
| In one phase | What it means | Phase score |
|---|---|---|
| 0 supports | Nothing but your feet took weight | 5 |
| 1 support | One hand, forearm, knee or leg-side | 4 |
| 2 supports | Two of them, in that phase | 3 |
| 3 supports | Three | 2 |
| 4 supports | Four | 1 |
| 5 supports | Five or more | 0 |
| A wobble | Visible loss of balance in that phase | −½ |
Both phases start at five, so the whole test starts at ten and every deduction is something that actually happened. The two phases are added, which is why halves exist on the scale and why a 7.5 is a real reading rather than a rounding of something else.
The rubric counts supports, not effort. A phase that was slow, ugly and unsupported scores five; a phase that looked composed with a hand on the knee scores four. That is a limit of the test worth knowing before you argue with your own number.
Sitting-rising test scores by age
There is no adult table, and the ones you will find are invented. Nobody has published age percentiles for the sitting-rising test in adults — not a mean for a 50-year-old, not a target for a 70-year-old. The two adult studies sorted people into score bands and followed the bands for years; producing a percentile was never what they were for.
The only age-stratified norms that exist are for children. de Oliveira Cunha and colleagues scored 596 Brazilian schoolchildren aged 5 to 14 and set reference percentiles by year of age, published in the Journal of Paediatrics and Child Health in 2026. They report performance in three bands — below expected, adequate, above expected — and found body mass index explained about 17% of the variation in the score, with no difference between boys and girls. Useful, and about children in one Brazilian state.
What the adult work does say about age is narrower and worth having. Both cohorts were middle-aged and older — 51 to 80 in 2014, 46 to 75 in 2025 — and within that range the score falls as age rises. Nobody has measured the test on adults younger than 46 in a study of this kind, so if you are under that age the mortality framing was never measured on people like you: take the movement reading and leave the rest of it alone.
The comparison that is available to everybody is the one against yourself. Score it today, write the number down with the date, and score it again in six months. A point lost between two of your own readings is a fact about you; a point above or below a table on the internet is a fact about the table.
What this reading is not
It is not a diagnosis. It is a movement score with a deduction sheet behind it. It does not name a condition, it does not rule one out, and it cannot tell you which joint, which muscle or which system produced the number.
It is not a lifespan calculator. The published figures are group rates over years of follow-up in one clinic’s population. Nothing about them converts into a date, a probability for a person, or a countdown, and any page that shows you one has invented it.
It is not worth much as a one-off if you are hurt. Taken with a painful knee, a stiff back on the day, or three days after a hard session, the score is a reading of that circumstance. One reading in that state is a data point about the day, not about you.
It is not a Limber score. The app’s five tests read 0–100 on its own anchors and roll into a Range Score. This one reads 0–10 on a published rubric and rolls into nothing. The two scales are not convertible and this page will not pretend they are.
If knee or hip pain is what stopped you — on the way down, at the bottom, or on the way up — the score is the least interesting thing about that attempt. See a clinician who can examine you rather than re-taking the test to see if it happens again.
Beside the battery, not in it
The five readings below are Limber’s battery, and they travel: whatever you have measured rides along in every link on this page, so you can walk the whole survey and arrive at the Range Score with it. The sitting-rising score is not in that list and never enters that composite. It is not one of the five, it feeds no area and no score group, and its scale is somebody else’s.
It does not travel either. This page’s own reading lives in this page’s address and nowhere else — copy the address before you walk on if you want to keep it.
0 of 5 tests taken. Every link here carries what you have already measured, so the readings travel with you from one instrument to the next.
Nothing on this site is stored and nothing is sent. Your readings live in the address bar and nowhere else, so a link is a survey and closing the tab loses it. Copy the address if you want to keep it, or take the same tests in Limber, where a reading is kept on the phone and measured against the one before it.
Stop at a stretch, never at sharp pain. Getting to the floor and back up loads knees, hips and the lower back at their end range, and a joint that catches, gives way or sends pain down a limb is not asking for another attempt.
See a clinician rather than a web page if a knee or hip locks, catches or gives way, if there is swelling, if a joint was injured recently, if you have fallen in the past year, or if you feel unsteady on your feet in ordinary life. This test does not diagnose anything and Limber does not treat anything.
How is the sitting-rising test scored?
Two phases, five points each, so the test starts at ten. Going down is scored out of five and coming up is scored out of five, separately. One whole point comes off for every support you put weight through in that phase — a hand, a forearm, a knee, the side of a leg, a hand pressed on your own knee, or a piece of furniture. Half a point comes off if you visibly lost balance in that phase. Add the two phases and that is the score, which is why halves exist on the scale and why 7.5 is a real reading. A phase stops at nought; the rubric has nothing below it. The form above does the subtraction and writes the arithmetic out under the reading so you can check it.
Is there a sitting-rising test score by age?
Not for adults, and the internet is full of tables that pretend otherwise. The only age-stratified norms anybody has published are for children: de Oliveira Cunha and colleagues scored 596 Brazilian schoolchildren aged 5 to 14 and set percentiles by year of age, published in the Journal of Paediatrics and Child Health in 2026. There is nothing equivalent for adults. The two adult studies grouped scores into bands and followed the groups for years — they never produced a percentile for a 55-year-old, and a page that shows you one has invented it. What the adult work does say is that the score falls with age across the range it studied, which was 46 to 80. If you are under 46, none of the mortality framing was measured on people like you.
What is a good sitting-rising test score?
The published work grouped scores rather than grading them, and the top group is 8 to 10: a 10 means you sat down and stood back up without a hand, a forearm, a knee or a leg-side taking any weight, and without wobbling in either direction. Below that, every whole point is one support you leaned on and every half point is a loss of balance. There is no population percentile here to compare yourself against — both published cohorts were volunteers at one exercise-medicine clinic in Rio de Janeiro, aged 46 to 80 and about two thirds men, which is not a sample of adults in general.
Does a low sitting-rising score mean I will die sooner?
No. The studies found an association in a group followed over years, not a prediction about a person. In the 2014 cohort of 2,002 adults aged 51 to 80, the group scoring 0 to 3 died at 5.44 times the rate of the group scoring 8 to 10 over a median 6.3 years of follow-up, adjusted for age, sex and body mass index. That is a rate in a group. It gives you no date, no probability for yourself, and no evidence that raising the score would change anything — nobody has run the trial that would answer that question.
What counts as a support in the sitting-rising test?
Anything that takes weight and is not your feet: a hand, a forearm, a knee, the side of a leg, or a hand pressed on your own knee. Leaning on furniture or a wall counts as well. Each support costs one point in the phase where it happened, sitting and rising being scored separately out of five, and a visible loss of balance in a phase costs half a point. Both phases start at five, which is why the whole test starts at ten.
Is the sitting-rising test a flexibility test?
Only in part, and its authors say so plainly: they describe it as reading muscular strength and power, flexibility, balance and body composition at the same time. That breadth is the reason a single number from it is interesting and also the reason it cannot tell you what produced the number. A score that falls between two attempts might be strength, might be balance, might be a stiff hip on the day, and the test cannot separate them.
Can I practise the test to improve my score?
You can practise getting to the floor and back up, and the score will very likely move, because it is a scored movement and movements respond to practice. What that does not tell you is whether anything else changed. The published association is between a score and an outcome in a cohort; nobody has shown that moving the score moves the outcome, and this page will not imply otherwise. Practise it because getting off the floor unaided is worth being able to do.
Should I take this test if my knees hurt or I feel unsteady?
It was published as a supervised assessment, with somebody watching and space to fail safely, and that is how to treat it. If a knee or hip hurts on the way down, if you have fallen in the past year, or if you feel unsteady on your feet in ordinary life, have someone with you or do not take it at all — not taking a test is a legitimate answer and a better one than a fall. Pain that catches, locks or runs down a limb is a question for a clinician who can examine you, not for a score.
The whole lower body as one screen — depth, heels and whether the bottom of the squat is somewhere you can rest.
Hip Mobility TestThe hips, in three positions: a squat hold and a hip flexor check scored on the app's own anchors, and a rotation compare that is read left against right rather than scored.
Knee to Wall TestAnkle dorsiflexion in centimetres, left and right kept apart. One of two tests in the battery that produce a real measured number.
Range ScoreThe composite. Whatever you have measured, rolled into one 0–100 reading and drawn as an elevation profile — with the ground you skipped left dashed.
- Brito LB, Ricardo DR, Araújo DS, Ramos PS, Myers J, Araújo CG. Ability to sit and rise from the floor as a predictor of all-cause mortality. European Journal of Preventive Cardiology 2014;21(7):892–8. doi:10.1177/2047487312471759
- Araújo CGS, de Souza e Silva CG, Myers J, Laukkanen JA, Ramos PS, Ricardo DR. Sitting-rising test scores predict natural and cardiovascular causes of deaths in middle-aged and older men and women. European Journal of Preventive Cardiology 2025. doi:10.1093/eurjpc/zwaf325
- de Oliveira Cunha C, Greco ALR, de Souza GHL, Guimarães LA, Bizinotto T, Porto CC, Formiga CKMR. Age-Specific Normative Values for the Sitting-Rising Test in Brazilian Children and Adolescents. Journal of Paediatrics and Child Health 2026. doi:10.1111/jpc.70507
Each address was followed to the record it names before it was written down. Where a paper could not be resolved that way, the page names it in the prose and leaves it unlinked rather than guessing at an address.
One reading is a point. Two are a line.
This page scores one test. Limber runs five — toe touch, knee-to-wall, shoulder reach, hip flexor and deep squat — rolls them into a single 0–100 Range Score, cuts that first score as your datum, and measures every later reading from it. The two wall tests are read in centimetres, the per-side tests keep left and right apart, and areas you never tested stay drawn as untested rather than filled in with a guess.
It works offline and needs no account: your readings stay on the phone.
Limber is in build. There is no store listing yet, so there is no download button here — when there is one it will be a link, not a promise.
This reading sits beside the battery, not inside it. Limber’s five tests compose into the Range Score; this one does not enter it, because it is not one of the five and its scale is not the app’s.
Nothing is scored here that the study did not score, and nothing about the app changes because of a number on this page.
Where the scoring comes from.
The 10-point rubric — five for sitting, five for rising, one off per support, half off for an unsteady phase — is de Brito and colleagues, European Journal of Preventive Cardiology, 2014, with the half-point deduction as stated in the same group’s 2025 follow-up. Both abstracts were read before this page was written.
Every other instrument in this case scores on Limber’s own anchors. This one does not, and that is the only reason it is allowed to score at all.
The rest of the case