What a vertical jump, a 1RM, and a skinfold actually measure, how each is supposed to be run start to finish, and the statistics — z-scores, percentiles, effect size, smallest worthwhile change — that turn a raw number into a defensible conclusion.
What actually shows up on the exam
Before a single stopwatch comes out, a tester has to know exactly what quality a given test is actually capturing — because several of these parameters look similar on the surface but are physiologically distinct, and the exam is built around exposing that confusion. A test is only useful if it's both reliable (gives a consistent result under the same conditions) and valid (actually measures the thing it claims to measure) — a test can be perfectly reliable and still be measuring the wrong quality entirely.
| Parameter | What it captures | A field test built around it |
|---|---|---|
| Strength | The maximal force a muscle or muscle group can generate in a single voluntary effort | 1RM back squat, bench press |
| Power | How quickly that force can be expressed — force multiplied by velocity, not force alone | Vertical jump, broad jump |
| Speed | How fast the whole body covers a fixed distance in a straight line | 40-yard / 40 m sprint |
| Agility | How fast the body changes direction, whether pre-planned or reacting to a stimulus | Pro agility (5-10-5), T-test |
| Aerobic capacity | How much energy the oxidative system can supply to sustain submaximal work over minutes | 1.5-mile timed run, multistage shuttle run |
| Anaerobic capacity | How much energy the phosphagen and glycolytic systems can supply during a short, near-maximal bout | Wingate test, repeated-sprint protocols |
| Muscular endurance | Ability to sustain repeated submaximal contractions before the muscle fails | Push-up test, timed curl-up test |
| Flexibility | Available range of motion at a single joint or a linked series of joints | Sit-and-reach |
| Balance | Ability to keep the body's center of mass controlled over its base of support | Y-balance test, single-leg stance |
| Body composition | The relative proportion of fat mass to fat-free mass | Skinfold measurement, bioelectrical impedance |
Where people lose points
A vertical jump is not a strength test — a heavier, stronger athlete can jump lower than a lighter, more explosive one, because jump height depends on how fast force gets applied, not just how much force is available. The same logic separates speed (a straight sprint) from agility (a sprint plus at least one forced change of direction) — a fast 40-yard time doesn't guarantee a fast pro agility time, and vice versa.
Aerobic and anaerobic capacity get confused for the same reason strength and power do: they sound like two points on a single continuum, but they're powered by different energy systems and respond to different training stimuli. A field test that hammers one — a long timed run for aerobic capacity, repeated all-out sprints for anaerobic capacity — is a poor stand-in for the other.
Administering a test well is mostly about controlling everything except the variable you're trying to measure. Same equipment, same setup, same verbal instructions and encouragement, same warm-up, same time of day where possible — any of those drifting between test and retest adds noise that can swamp a real change in performance. Two ideas sit underneath every protocol below.
| Concept | Question it answers | What tends to break it |
|---|---|---|
| Reliability | Same test, same athlete, same day-to-day conditions — same result? | Different technicians, inconsistent cueing, equipment that drifts out of calibration, athlete fatigue |
| Validity | Is this test even measuring the parameter it's supposed to? | Choosing a test that's easy to run but doesn't isolate the target quality (e.g., using a max-effort long run to represent "speed") |
Test order within a single session matters for the same reason: a fatiguing test performed early will quietly degrade every test that follows it. The general convention is to run tests roughly in order of how much residual fatigue they leave behind — resting measures and non-fatiguing assessments first, then agility and speed, then power, then maximal strength, then anaerobic and finally aerobic capacity testing, since the aerobic test tends to leave an athlete too fatigued to produce a valid effort on anything else.
| Test | What it targets | General procedure |
|---|---|---|
| Vertical jump (jump-and-reach) | Lower-body power | Standing reach height is marked first with the arm fully extended overhead; the athlete then performs a maximal-effort jump (countermovement allowed) and touches or marks the highest point reached; jump height is the difference between the two marks; the best of several attempts is recorded |
| 1RM test | Maximal strength | Athlete warms up at a comfortable submaximal load, then attempts progressively heavier single lifts with full recovery (roughly 3–5 minutes) between tries, aiming to land on a true one-rep max within about 3–5 total attempts; spotters and strict technique are required throughout |
| 40-yard / 40 m sprint | Straight-line speed | Athlete starts from a stationary two- or three-point stance just behind the start line and sprints maximally through the finish without decelerating early; timed with a stopwatch or, more precisely, electronic timing gates; best of multiple trials is used |
| Pro agility / 5-10-5 shuttle | Change-of-direction speed | Athlete straddles a center line, sprints 5 yards to touch one side line, reverses direction and sprints 10 yards to touch the far line, then sprints back 5 yards through the start line; timed from the first movement to crossing the finish |
| Sit-and-reach | Hamstring and low-back flexibility | Seated with knees fully extended and feet flat against a box, the athlete reaches forward along a ruled scale as far as possible without bouncing and holds the farthest point briefly; best of a few trials is recorded |
| Y-balance test | Dynamic single-leg balance | Standing on one leg on a center platform, the athlete reaches the free leg as far as possible along three lines (anterior, posteromedial, posterolateral) without losing balance or touching down for support; reach distances are normalized to leg length so athletes of different heights can be compared |
| Skinfold / bioelectrical impedance (BIA) | Body composition | Skinfold: calipers pinch a double fold of skin and subcutaneous fat at standardized anatomical sites, and the sum of sites is run through a population-specific equation to estimate percent body fat. BIA: a low-level electrical current is passed through the body, and the resistance it meets — driven mainly by tissue water content — is used to estimate fat-free mass |
These three parameters share a common vulnerability: their tests are more sensitive to measurement technique than most strength or speed tests are, which means the exam likes to test what can quietly go wrong with each one.
Applied — why the same technician matters
Body composition testing is rarely used to declare an athlete's "true" body fat percentage — every field method carries some error against a lab reference like DEXA. What it's actually good for is tracking change within one person over time, and that only works if the source of error stays constant between measurements. Swap technicians, sites, or the time of day, and you've introduced a new variable that can look exactly like a real change in body composition when it isn't.
A raw score by itself doesn't say much — 27 inches on a vertical jump is meaningless without something to compare it to. Statistics are what turn that number into a usable conclusion: where does this score sit relative to a group, and is a change from one test session to the next big enough to actually mean something.
| Statistic | What it is | Why it matters |
|---|---|---|
| Mean | The arithmetic average of a set of scores | The reference point every other statistic below is built around |
| Standard deviation (SD) | The average distance individual scores sit from the mean | Describes spread — a small SD means scores cluster tightly; a large SD means they're scattered widely |
| Z-score | (Individual score − Mean) ÷ SD | Converts a raw score into "how many standard deviations from average," which lets scores from completely different tests and units be compared directly |
| Percentile rank | The percentage of a reference group's scores that fall below a given score | Translates a z-score into a more intuitive statement — "this score beat X% of the comparison group" |
The sign-flip trap
The z-score formula assumes higher is better. On any test where a lower number is the better outcome — a 40-yard sprint, a pro agility time — a raw z-score has to be interpreted in reverse, or the sign flipped, before it means what people expect it to mean. An athlete who runs faster than average will produce a negative raw z-score on time, even though their performance is above average. Skipping that flip is one of the most common scoring errors when building a multi-test athletic profile.
Percentile rank and z-score describe the same information two different ways — one is a standardized distance from the mean, the other is a share of a comparison group beaten. Neither one, by itself, says whether a change between two test sessions is real or just noise. That's what effect size and smallest worthwhile change are for.
| Cohen's d | Label | What it means practically |
|---|---|---|
| < 0.2 | Trivial | Essentially no meaningful practical difference, even if a statistical test calls it "significant" |
| ~0.2 | Small | A real but modest difference — easy to overlook without formal measurement |
| ~0.5 | Medium | A difference a careful, experienced observer would likely notice without needing to measure it |
| ≥ 0.8 | Large | A difference substantial enough that most people would notice it without any formal testing at all |
Cohen's d itself is calculated as the difference between two means divided by a pooled standard deviation — structurally the same idea as a z-score, just applied to the gap between two groups (or two time points) instead of one score against a distribution.
Smallest worthwhile change (SWC) — is the change real, or just noise?
Every test carries some amount of built-in measurement error — even a perfectly performed retest will bounce around slightly from session to session just from normal biological variability. Smallest worthwhile change is the threshold a score has to cross before it's considered a meaningful, practically important improvement rather than random noise; a common approach sets it around 0.2 × the between-athlete SD, mirroring the "small effect" threshold from Cohen's d.
For that threshold to be useful, the test's own noise — often reported as a typical error of measurement or a coefficient of variation — has to be smaller than the SWC. A test too noisy to reliably detect a change of that size isn't sensitive enough to use for tracking that particular improvement, no matter how valid it is in principle.
A single test score describes one quality. An athletic profile combines several tests — say, vertical jump, 1RM squat, 40-yard sprint, and pro agility — into one picture of an athlete's overall strengths and weaknesses, and that only works if the scores are put on the same scale first.
A shortcut for the whole statistics section
Z-score answers "how far from average, in SD units." Percentile restates that same answer as "beats X% of the group." Effect size answers "how big is the gap between two means, in SD units" — same math, different question. Smallest worthwhile change answers "is that gap big enough to actually matter." Four different questions, one shared building block: the standard deviation.
A vertical jump test is most directly a measure of which performance parameter?
C — power. Jump height reflects how quickly force can be applied (force × velocity), not the maximum force available in isolation. A stronger athlete isn't guaranteed to jump higher than a more explosive one.
During a 1RM protocol, how many total attempts should a tester generally aim to land a true maximum within?
B. After an initial warm-up set, load is increased across single attempts with full recovery between them, with the goal of finding the true max within roughly 3–5 total attempts — enough to close in on the ceiling without accumulating fatigue that would compromise the result.
A pro agility (5-10-5) test is best classified as measuring which parameter?
B. The 5-10-5 requires two full changes of direction under time pressure, which is the defining feature that separates agility from a straight-line speed test like a 40-yard sprint.
A group's vertical jump has a mean of 22 inches and a standard deviation of 3 inches. An athlete jumps 28 inches. What is that athlete's z-score?
B. z = (28 − 22) ÷ 3 = 6 ÷ 3 = +2.0 — two full standard deviations above the group mean, which corresponds to roughly the 98th percentile.
On a 40-yard sprint test, an athlete's raw z-score for time comes out to −1.5. What does that actually mean about their performance?
B. Because a lower time is the better outcome on a sprint test, a negative raw z-score on time actually represents above-average (faster) performance — the sign has to be interpreted in reverse compared to a "higher is better" test like vertical jump.
An athlete's power score sits at the 84th percentile compared to a normative database. What does that mean?
A. The 84th percentile corresponds to roughly a +1.0 z-score, and means the athlete's score exceeded 84% of the comparison group's scores.
Two training groups' pre-to-post strength gains are compared and Cohen's d works out to 0.65. How should that effect size be classified?
C — medium. 0.65 falls between the roughly 0.5 medium threshold and the roughly 0.8 large threshold, closer to medium.
A retest shows a statistically significant improvement in an athlete's jump height, but the change is smaller than that test's smallest worthwhile change. What's the most defensible conclusion?
B. Statistical significance and practical significance are separate questions. A change that's technically detectable but smaller than the smallest worthwhile change may simply reflect normal test-retest noise rather than a real training effect.
Which factor is most likely to introduce error into a skinfold-based body composition estimate across two testing sessions?
A. Skinfold accuracy depends heavily on consistent site location and pinch technique; changing the technician or site markings between sessions introduces a source of error that can look identical to a real change in body composition.
When building a multi-test athletic profile out of tests with different units (inches, seconds, pounds), what has to happen before the scores can be compared on one chart?
B. Raw scores in different units aren't directly comparable. Standardizing every test to a z-score — and flipping the sign on tests where a lower raw number is the better result — puts every dimension of the profile on the same "higher is always better" scale.
This chapter is original study material covering topics also addressed in Essentials of Strength Training and Conditioning, 5th edition (NSCA/Human Kinetics). It is an independent work, not a reproduction or paraphrase of that text. This guide is not affiliated with, endorsed by, or sponsored by the National Strength and Conditioning Association (NSCA). CSCS® and Certified Strength and Conditioning Specialist® are registered trademarks of the NSCA.