How to Measure Knee Flexion With a Goniometer

Measuring knee flexion with a goniometer comes down to three landmarks and careful alignment: you place the goniometer’s pivot point on the outer side of the knee, aim one arm up toward the hip, and aim the other arm down toward the ankle. The technique is straightforward in concept but surprisingly easy to get wrong in practice, and even small positioning errors can shift your reading by ten degrees or more. Whether you are tracking recovery after surgery, monitoring rehab progress, or screening an athlete, getting accurate and repeatable numbers depends on understanding where things tend to go sideways.

The Three Landmarks That Matter

A standard universal goniometer has a stationary arm, a moving arm, and a central pivot (the axis). For knee flexion, you align the axis at the lateral femoral epicondyle, which is the bony bump on the outer side of the knee. The stationary arm points toward the greater trochanter, the prominent bone at the outer hip. The moving arm follows the lower leg down toward the lateral malleolus, the bony knob on the outside of the ankle.¹ Some clinical protocols use the proximal head of the fibula instead of the epicondyle as the axis landmark, and the lateral malleolus as the distal reference for the moving arm.² The differences between these two setups are small in practice, but consistency matters: pick one protocol and stick with it across all your measurements on a given patient.

Before taking the measurement, have the person bend their knee to the end of the range you are testing. For active flexion, they do the bending themselves. For passive flexion, you gently push the lower leg toward the thigh until you feel firm resistance. Keep the goniometer flat against the skin, making sure the arms stay aligned with the long axes of the thigh and the lower leg. Read the angle where the moving arm meets the protractor scale. Full knee extension is considered zero degrees, and flexion increases from there.

Does Patient Position Change the Reading?

One of the most common questions is whether it matters if the person is lying on their back, sitting upright, or in some other position. The short answer: for most clinical purposes, no. A study of patients after total knee replacement found no significant differences between measurements taken in three different body positions, and the readings across those positions were strongly correlated.³ That said, sitting generally produces tighter agreement between different examiners than supine lying does. One study found that passive flexion measured in a sitting position had narrower limits of agreement between two observers compared with supine, where the gap between examiners ranged from about minus 16 to plus 19 degrees.⁴

There is one important caveat here: hip position can influence knee flexion readings. When the hip is extended, the rectus femoris muscle, which crosses both the hip and the knee, pulls tighter and can limit how far the knee bends. Research on patients undergoing knee replacement found that hip extension significantly reduced knee flexion angle, and this effect was even more pronounced after surgery.⁵ If you are measuring someone who is lying flat on their back with their leg straight, the hip is in relative extension, and you might get a lower flexion number than you would if they were sitting with the hip flexed. For tracking progress over time, the practical takeaway is simple: measure the same person in the same position every time.

How Reliable Are Goniometer Measurements, Really?

Goniometry gets a reputation as imprecise, and there is some truth to that, but the numbers are better than many people assume. When the same clinician measures the same knee repeatedly, reliability is high. A study comparing the universal goniometer, a digital inclinometer, and a smartphone app in knee replacement patients found that all three tools showed high reliability both within and between raters, with intraclass correlation coefficients ranging from about 0.75 to 0.95.⁶ Another study found concordance correlation coefficients above 0.98 for repeated knee flexion measurements with a universal goniometer, and the reliability held up even when novice clinicians did the measuring rather than experienced ones.⁷

Where things get shakier is inter-rater reliability, meaning how closely two different people agree when measuring the same knee. The measurement error between clinicians is usually larger than the error within one clinician. This is not a flaw of the goniometer itself so much as a reality of how differently people palpate landmarks, apply pressure, and read scales. One study found that experienced physiotherapists systematically recorded larger range-of-motion values than inexperienced ones, likely because of differences in physical size, strength, and how much force each applied during passive measurements.⁸ Body composition of the patient also plays a role. Overweight patients have more soft tissue between the goniometer arms and the bony landmarks, making precise alignment harder.

Choosing the Right Goniometer

The standard “long-arm” universal goniometer, with arms roughly 30 centimeters long, is the workhorse of clinical practice. Shorter-arm models are more portable and easier to handle, but the tradeoff is accuracy. Long-arm goniometers produce better intra- and inter-observer reliability than short-arm versions. In one direct comparison, the long-arm goniometer achieved an intraobserver ICC of 0.92 and an interobserver ICC of 0.88, while the short-arm model came in at 0.78 and 0.56 respectively.⁹ The reason is geometric: longer arms make small alignment errors less impactful on the final angle reading, the same reason a long ruler is easier to draw a straight line with than a short one.

A study that directly compared five goniometric methods found that a digital inclinometer had the smallest minimum significant difference at about 6 degrees, meaning a change of at least 6 degrees was needed to be confident the knee had truly moved more. The long-arm goniometer required a 10-degree change, a smartphone app needed 12 degrees, and both visual estimation and the short-arm goniometer required 14 degrees.¹⁰ Those minimum-significant-difference thresholds matter a lot in rehab: if your tool cannot reliably detect a change smaller than 14 degrees, you might miss real progress or falsely conclude that a patient is improving when they are not.

Fluid goniometers, which use a bubble level mechanism rather than a pivot and arms, are another option. They work well in some joints but can be tricky for knee flexion. The fluid level is sensitive to small tremors in the lower leg when the patient holds a flexed position against gravity, and this can introduce substantial measurement variability.¹¹ For most situations, sticking with a standard long-arm goniometer or a validated digital tool is the safer bet.

Smartphone Apps as an Alternative

If you do not have a goniometer handy, a smartphone can actually do a decent job. Several validated apps use the phone’s built-in accelerometer and gyroscope to measure joint angles. A study comparing a smartphone goniometer app to a traditional universal goniometer reported Pearson correlations of 0.96 to 0.98 and intraclass correlation coefficients of 0.97 to 0.99 for individual knee joint angles.¹² Another study comparing a smartphone app against computer-navigated measurements taken during surgery found a mean difference of only about 1 degree between the two methods, with good intra-observer and inter-observer reproducibility.¹³

There is even some evidence that smartphone apps produce smaller measurement error than a traditional goniometer. One study found that the standard error of measurement for knee flexion was about 2.7 degrees with a phone app compared with 3.4 degrees with a universal goniometer, and the minimum detectable change was about 6.3 degrees versus a larger value for the manual tool.¹⁴ Phone apps are not perfect, and the minimum significant difference in one comparison was 12 degrees, worse than a digital inclinometer or long-arm goniometer. But for home monitoring or when a traditional goniometer is not available, a well-validated app is a reasonable substitute.

Common Sources of Error and How to Avoid Them

Most measurement mistakes fall into a handful of predictable categories. Knowing them helps you get more consistent readings.

Landmark drift: The bony landmarks you palpated at the start of the measurement can shift under your fingers as the knee bends, especially in people with more soft tissue. Re-check your landmarks at the end-range position before recording the number.
Axis misplacement: If the goniometer’s pivot is not centered over the joint’s axis of rotation, the angle will read high or low depending on which direction you are off. The lateral epicondyle is the standard target, but it can be hard to find in a swollen knee.
Arm misalignment: The stationary and moving arms need to track the long axis of the femur and tibia respectively. If one arm drifts anterior or posterior, the reading will be off. Using a long-arm goniometer reduces this error because slight deviations are proportionally smaller over a longer span.
Inconsistent force: During passive flexion measurement, how hard you push matters. Two clinicians applying different amounts of pressure will get different end-range angles. Standardize your end-feel: push until you meet firm resistance, not until the patient winces.
Swelling and pain: A knee that is swollen or painful will not bend as far. This is a real limitation in the joint, not a measurement error per se, but it means your number reflects the knee’s current state rather than its structural capacity. Note the presence of effusion when you record the measurement.

One factor that does not seem to matter much is time of day. Research on clinical screening tests found that time of day affected hip joint position sense but did not significantly affect knee joint measurements.

What the Numbers Mean for Daily Life

Knowing that a knee bends to 110 degrees is only useful if you know what 110 degrees lets you do. Research using electrogoniometry during real activities found that walking on flat ground and going up or down slopes requires less than 90 degrees of knee flexion. Stairs and getting in and out of a chair require roughly 90 to 120 degrees. Getting into a bathtub needs approximately 135 degrees.¹⁵ That study suggested about 110 degrees as a reasonable rehabilitation target for functional independence after knee surgery.

A more recent study specifically looking at stair-climbing ability one year after bilateral knee replacement found a cutoff of about 121 degrees of knee flexion for the ability to independently go up and down stairs.¹⁶ The gap between 110 and 121 degrees reflects how the question is framed. The earlier number is a practical minimum for getting through most daily activities, while the stair-specific cutoff accounts for the deeper bend needed to manage steps comfortably. Quadriceps strength matters too, so range of motion alone does not tell the whole story about function.

For people recovering from ACL reconstruction rather than knee replacement, the benchmarks are different. The goal is typically to match the uninjured side. Clinical criteria from the International Knee Documentation Committee consider normal knee flexion to be within 5 degrees of the opposite knee, and normal extension to be within 2 degrees.¹⁷ The emphasis here is on symmetry rather than hitting a fixed number, because people’s baseline ranges vary quite a bit from one individual to the next.

Tracking Recovery Over Time

One of the most valuable uses of goniometry is not any single measurement but the trend over weeks and months. After knee replacement, for instance, flexion typically follows a predictable curve: a rapid gain in the first few weeks, then a slower climb over several months. Researchers have developed reference charts that plot the expected recovery trajectory along with the normal range of variability around it, so that a clinician can compare a specific patient’s progress against what is typical for their stage of recovery.¹⁸ This approach moves away from one-size-fits-all milestones and toward individualized decision-making.

For tracking to work, measurement consistency is everything. As covered earlier, the same person should measure the same patient in the same position with the same tool each time. If you switch from a short-arm goniometer to a phone app, or from supine to sitting, any apparent change could be an artifact of the method rather than real progress in the joint. This is especially important in the early postoperative period, when swelling and pain fluctuate day to day and can mask or mimic actual gains in range of motion.

Active Versus Passive Range of Motion

When you record a knee flexion angle, you should note whether it was measured actively or passively. Active range of motion is what the patient can achieve under their own muscle power. Passive range of motion is what you can achieve by manually bending the knee for them. Passive numbers are almost always higher than active ones, because muscle weakness, pain inhibition, or swelling can prevent a person from pulling the knee through its full available range.

Interestingly, from a pure joint-mechanics standpoint, the motion path of the knee does not differ significantly between active and passive movement. An intraoperative study using navigation technology during knee replacement surgery found no statistically significant differences in the rotational and translational kinematics of the knee between active and passive flexion-extension, either before or after implant placement.¹⁹ The joint itself tracks the same way; the difference between active and passive range comes from the soft tissues and muscles around it, not from the joint surfaces. That means when a patient’s active flexion is noticeably less than their passive flexion, the bottleneck is usually muscle strength or pain rather than a structural block inside the joint. This distinction helps clinicians choose the right intervention: strengthening exercises if the gap is large, versus manual therapy or surgical review if passive range is also limited.

Measuring Children

Most of the normal values and benchmarks discussed so far come from adult populations, and applying them to children is a mistake. Joint range of motion in children changes with age, and a study of typically developing children between ages 4 and 16 found a clear correlation between joint range and age across most lower-limb measures.²⁰ Younger children tend to have greater flexibility overall, and what looks like excessive motion in an adult might be completely normal in a six-year-old. If you are measuring a child’s knee flexion, you need age-matched reference values rather than adult norms. The goniometric technique itself is the same, though smaller landmarks and a less cooperative patient can make alignment trickier. Using a long-arm goniometer on a very small child can feel unwieldy, but downsizing to a short-arm model trades away the accuracy advantage of longer arms.

When Visual Estimation Falls Short

Plenty of clinicians glance at a knee and estimate the angle without pulling out a goniometer. In a busy clinic, this is understandable. But the data is not kind to eyeballing. Visual estimation and short-arm goniometry were equally inaccurate in a head-to-head comparison, both requiring a 14-degree change before you could be confident the difference was real.¹⁰ That is a meaningful margin. A patient might go from 95 degrees to 108 degrees of flexion over two weeks and, by visual estimation alone, a clinician could genuinely miss the improvement entirely. For documentation, insurance authorization, or any situation where the number will be compared against a benchmark or a previous measurement, take 30 seconds and use an actual instrument.

This is where digital tools earn their keep. A digital inclinometer, with its 6-degree minimum significant difference, can detect real changes roughly twice as sensitively as visual estimation. If your practice sees a lot of post-surgical knees or ACL rehab patients, investing in a digital inclinometer or using a validated phone app is a meaningful upgrade in measurement quality for very little extra effort per patient visit.