Effect Size (Cohen's d) Calculator
The formula
What the cohen's d tells you
The mean difference expressed in standard deviations — effect magnitude without the units, comparable across any measurement scale.
Calculation: Cohen's d = (M₁ − M₂) ÷ pooled SD. Enter your inputs and every substitution is shown below the result.
Reading the result
d turns “significant” into “how much”: p tells you an effect exists; d tells you whether anyone should care.
Rules of thumb: 0.2 / 0.5 / 0.8 = small / medium / large convention
Limitations and best practice
Cohen’s small/medium/large labels were meant as rough defaults, not field norms — in drug efficacy, d=0.2 is a blockbuster.
This calculator gives an estimate based on standard formulas; it is not professional financial, medical, legal or engineering advice. Verify anything you plan to act on with a qualified professional and your own records.
Frequently asked questions
Which SD for pooling?
Does d fix the multiple-comparisons problem?
Source & methodology
Cohen (1988); Hedges & Olkin meta-analysis conventions.
Effect Size (Cohen's d): what this tool answers
This page was written for one job: turning your own inputs into a defensible figure, with the working shown. There are plenty of pages about Effect Size (Cohen's d); almost none of them show the working. This one does: the formula, every input, a fully computed example, and a sensitivity analysis calculated live from this engine’s own defaults.
The useful way to hold Effect Size (Cohen's d) is as a ratio with a story attached: Cohen's d = (M₁ − M₂) ÷ pooled SD says how the parts combine, and the 0.43 figure above is simply that sentence written in numbers. Everything that follows unpacks one clause of the sentence at a time.
In plain terms, Effect Size (Cohen's d) answers the question "Cohen's d?". The measurement matters because it converts a situation that feels complicated — several inputs pulling in different directions — into a single number you can act on. Within Science & Research, Statistics & Data Science is the context where this figure is quoted most often, and professionals treat it as a first-pass screen rather than a verdict.
Every figure in the prose is generated by the engine rather than written beside it — the worked example resolves to cohen's d = 0.43 on the default inputs, and the sensitivity section ranks the inputs that move the result — Group 1 mean first, at ≈126.7 % of the headline for a ten-percent error either way in that input — which is the thread the rest of this guide pulls on.
The Effect Size (Cohen's d) formula, explained
One formula, published in full: Cohen's d = (M₁ − M₂) ÷ pooled SD. Anything you can read, you can reproduce in a spreadsheet, and that reproducibility is the whole point of printing it rather than hiding it behind a button.
Group 1 mean (example value 76) is a measured number in the unit named by the label; the formula does not convert for you. This input feeds the formula directly; a defensible figure here is worth more than false precision anywhere else.
Group 2 mean (example value 70) is a quantity rather than a category — precision here is cheap and worth having. If you are unsure what belongs in this field, look at the printed steps: the substitution shows exactly how the engine uses it.
Pooled standard deviation (example value 14) is a plain measured quantity — the unit shown in the label is the unit the formula expects. If you are unsure what belongs in this field, look at the printed steps: the substitution shows exactly how the engine uses it.
The default Effect Size (Cohen's d) calculation, written out in full
The tool prints its own working; this is the same arithmetic on the example values, reproduced so you can follow it by hand. Starting figures: Group 1 mean, Group 2 mean, Pooled standard deviation. The endpoint of that chain is 0.43 of cohen's d, sitting in the “Small effect — real but subtle” band.
- d = (76.0 − 70.0) ÷ 14.0 = 0.43
- Ladder: 0.2 small · 0.5 medium · 0.8 large (Cohen’s conventions, not physics)
Percent-overlap interpretation (U₁ proxy) — 61.9 %. If your inputs change, this is usually the first companion figure to move.
n per group for 80% power at α=.05 — 88. Read next to the headline, this is the figure that shows the shape of the trade-off.
Effect in SD units — 0.43 SD. This is the number experienced users quote when the headline alone would mislead.
Difference raw — 6.0. Read next to the headline, this is the figure that shows the shape of the trade-off.
Keep this worked example open while you enter your own figures: matching the shape of the substitution is the fastest way to spot a unit or convention mistake.
Scenario table: Effect Size (Cohen's d) at ±10 % and ±20 % of the key input
The honest version of any estimate is a corridor, and this one is cheap to produce: Effect Size (Cohen's d) was recomputed at 5 points along the dominant input, with every other field left untouched.
| Group 1 mean setting | Cohen's d |
|---|---|
| 80 % of the example value | -0.66 |
| 90 % of the example value | -0.11 |
| example inputs | 0.43 |
| 110 % of the example value | 0.97 |
| 120 % of the example value | 1.51 |
That spread — -0.66 at the low end, 1.51 at the high — is what a careful read of Effect Size (Cohen's d) looks like: a band you can state out loud, with the reason for each edge. Reproduce it in the form above by editing Group 1 mean alone; the substitution steps will follow your figures exactly.
What actually moves your Effect Size (Cohen's d)
Generic advice says “it depends on your inputs”; this section says exactly how much. Each field was perturbed by ±10 % in isolation and the cohen's d recomputed, so the ranking below is measured rather than asserted. Read it as a map of where Effect Size (Cohen's d) is fragile — drawn for the example inputs, which is the map you want when deciding what to measure carefully.
- Nudge Group 1 mean by ±10 % and Effect Size (Cohen's d) lands between -0.11 and 0.97 (sensitivity ≈126.7 %)
- Nudge Group 2 mean by ±10 % and Effect Size (Cohen's d) lands between -0.07 and 0.93 (sensitivity ≈116.7 %)
- Nudge Pooled standard deviation by ±10 % and Effect Size (Cohen's d) lands between 0.39 and 0.48 (sensitivity ≈11.1 %)
Group 1 mean is the dominant lever here: ten percent either way moves the result by about 126.7 %, from -0.11 to 0.97. That is where a careless figure does the most damage, so check it against a source, and Group 2 mean is effectively tied with it (≈116.7 % for the same nudge), so verify the pair together. The remaining fields can sit at safe defaults without materially changing the answer — knowing which is which is what separates a careful calculation from an anxious one.
What an Effect Size (Cohen's d) of 0.43 means in practice
The value above only becomes useful the moment you know what it is being measured against. The default case lands in the “Small effect — real but subtle” band, and bands are deliberately wide because real situations overlap. The sensitivity list shows how far a modest input error moves you along that ladder: for Group 1 mean, ±10 % covers -0.11 to 0.97. The practical reading rules: compare against your own history before comparing against other people; treat band boundaries as zones, not walls — one point on the wrong side of a threshold rarely changes a real outcome; and when a companion output contradicts the headline, believe the pair.
Placing this figure in context: calculators of this kind differ less in arithmetic than in what they choose to ignore. This one keeps Group 1 mean explicit so that the omissions are visible rather than hidden in a preset.
How people get Effect Size (Cohen's d) wrong
Cohen’s small/medium/large labels were meant as rough defaults, not field norms — in drug efficacy, d=0.2 is a blockbuster. This calculator gives an estimate based on standard formulas; it is not professional financial, medical, legal or engineering advice. Verify anything you plan to act on with a qualified professional and your own records.
Copying a single headline number without checking the companion outputs. The tool reports Percent-overlap interpretation (U₁ proxy), n per group for 80% power at α=.05 and Effect in SD units alongside the main result precisely because Effect Size (Cohen's d) in isolation misleads — the secondary columns are where outliers and trade-offs show up.
Attention is also misallocated between fields. Having measured this engine, we can say where it belongs: Group 1 mean first, everything else second.
A final trap is presentation: reporting Effect Size (Cohen's d) to more decimal places than the inputs support. The tool prints what the arithmetic gives; you should quote only what your least precise field justifies.
A quick check for any Effect Size (Cohen's d) figure
- Verify each input against its original source document, not memory.
- State, in one sentence, which convention each rate, unit or period in the form is using.
- Re-verify Group 1 mean against a source document — ten percent of error there moves the headline by ≈126.7 %.
- Recompute once with the dominant input deliberately set low, and see whether the decision changes.
- Compare the result with one independent method or published benchmark.
- Repeat the ±10 % test on the inputs that matter and see whether the decision — not just the number — still holds.
Step by step: getting a trustworthy Effect Size (Cohen's d) out of this tool
Using a calculator well sounds trivial, but the difference between a decision-grade figure and a throwaway one lives entirely in the routine. The routine for Effect Size (Cohen's d) takes a minute to learn and applies to every input above.
Step 1 — start from evidence. The inputs behind Effect Size (Cohen's d) are group 1 mean, group 2 mean, pooled standard deviation. Write them down with their source next to each one; if you cannot name where a figure came from, that is the input most likely to be wrong. The inputs are the part of the work that nobody else can do for you.
Step 2 — start from the case you can defend. Enter pessimistic-but-honest figures first; the optimistic run is a ceiling, not a plan. A conclusion that holds on conservative inputs holds everywhere. Rounding early is the quietest way to introduce a permanent error.
Step 3 — change one input at a time. Nudge a single field by ten percent, watch the headline and its companions move, then undo it. Two or three fields will dominate — the sensitivity table above already ranks them for this engine. One decimal place more than your data supports is one decimal place of fiction.
Step 4 — stress before you commit. A decision that survives a ten-percent squeeze on its two biggest inputs needs no luck to be sound. One that does not survive is a decision about hope, not about arithmetic. Say the answer out loud before you save it; if it sounds wrong, the model did not change, your understanding did.
Step 5 — save the story with the figure. A number without its inputs and its date is unusable within a quarter. Record 0.43, the assumptions behind it, and today’s date; your future self will be auditing a stranger’s arithmetic. The test of a good estimate is not that it feels right but that it changes what you do.
The real-world conversations Effect Size (Cohen's d) settles
In Statistics & Data Science the figure computed here is usually the first one somebody asks to see. Having it — with the formula behind it — changes your position in the conversation from participant to reference. The point of transparency is that a reader can disagree with a specific line rather than with you.
The audience for this page is people who need one trustworthy number before their next decision. In professional Statistics & Data Science practice the pattern is always the same — screen with Effect Size (Cohen's d), then invest slow analysis only in whatever survives the screen. Half of a good calculation is knowing which input you would hate to be wrong about.
Treat Effect Size (Cohen's d) like a smoke alarm rather than a painting — its value is in being tested periodically, not in being admired once. The input you reach for first is usually the one you measured longest ago.
People also ask about Effect Size (Cohen's d)
What affects Effect Size (Cohen's d) the most?
Measured on this engine, Group 1 mean is the dominant lever: moving it ±10 % shifts the cohen's d across -0.11 to 0.97 — about 126.7 % of the headline, with Group 2 mean next at ≈116.7 %. The ranked list in the sensitivity section is computed live, not asserted from a template.
Do group 2 mean and group 1 mean matter equally?
Effect Size (Cohen's d) here is co-driven: Group 1 mean and Group 2 mean produced ≈126.7 % and ≈116.7 % responses, which is a tie for practical purposes. Everything below those two is a rounding error by comparison.
What is a typical Effect Size (Cohen's d)?
On the worked example above Effect Size (Cohen's d) comes out at 0.43. Treat that as one point on a distribution rather than a norm: the bands, the companion outputs and the sensitivity ranking tell you how wide the plausible range really is.
How do I calculate Effect Size (Cohen's d) by hand?
Cohen's d = (M₁ − M₂) ÷ pooled SD is the entire method. Work the inputs in that order — d = (76.0 − 70.0) ÷ 14.0 = 0.43 — and you will reach 0.43 on the default figures, the same number this tool prints.
What happens to Effect Size (Cohen's d) if group 1 mean changes by 10 %?
The engine was re-run with that field moved by ten percent while everything else stayed fixed: the cohen's d goes across -0.11 to 0.97, a swing of roughly 126.7 % around the 0.43 baseline. That ratio is the practical definition of how much this input matters.
Is the Effect Size (Cohen's d) result of 0.43 good or bad?
Good and bad are comparisons, so the honest test is three questions: where does your own Effect Size (Cohen's d) sit against the 0.43 reference above, how far would a plausible input error move it (≈126.7 % per ten percent on Group 1 mean), and what does the alternative option compute to?
How do I calculate Effect Size (Cohen's d) manually?
Write down the inputs, then follow the printed steps in order — Cohen's d = (M₁ − M₂) ÷ pooled SD. A spreadsheet replicates it with two or three cells. The manual route is worth doing once to build intuition; after that the calculator saves time and eliminates transcription slips.
How many significant figures are reliable in Effect Size (Cohen's d)?
About as many as your least precise input. The tool computes at full precision and displays rounded output; report 0.43 to a decimal place your inputs can actually support.
Which SD for pooling?
The pooled within-group SD (or pre-intervention SD for pre/post designs); using the total SD of a changed sample inflates d.
Does d fix the multiple-comparisons problem?
No — it sizes each effect honestly; each test’s false-positive risk still needs correction.
Does this Effect Size (Cohen's d) tool store my numbers?
No. The entire calculation runs in your browser session; nothing entered above is transmitted or saved, which is the correct behaviour for sensitive Science & Research figures and the reason you can re-run this page as often as you like.
What the labels on this page mean
- Percent-overlap interpretation (U₁ proxy)
- 61.9 % for the Effect Size (Cohen's d) worked case; it sits next to the main result because it carries context the headline leaves out.
- n per group for 80% power at α=.05
- 88 for the Effect Size (Cohen's d) worked case; it sits next to the main result because it carries context the headline leaves out.
- Effect in SD units
- 0.43 SD in the default Effect Size (Cohen's d) calculation — one of the secondary figures an experienced reader checks before trusting the primary number.
- Cohen's d
- The primary output of this tool: 0.43 for the default inputs, computed exactly from Cohen's d = (M₁ − M₂) ÷ pooled SD.
- Worked example
- Every substitution in order, from the example inputs to 0.43 — the audit trail for the headline figure.
- Verdict band
- The interpretation attached to a result (“Small effect — real but subtle” on these inputs), which turns a bare number into a position on a scale.
- Sensitivity (±10 %)
- How far the Effect Size (Cohen's d) moves when a single input is nudged ten percent either way; the ranked list on this page is computed live from this engine.
Where the Effect Size (Cohen's d) model comes from
Everything above runs on one stated formula, applied without hidden coefficients: Cohen's d = (M₁ − M₂) ÷ pooled SD. Source of record: Cohen (1988); Hedges & Olkin meta-analysis conventions. Any other implementation of this model should return the same figure for the same inputs. If one does not, the difference is in its assumptions — which this page prints and that one probably does not.
This page is an estimation aid, not professional advice. The design rule is simple: two people entering identical figures must get identical answers. The promise to the reader is narrow and testable: same inputs, same output, no silent rounding, and no figure presented without the formula that produced it.
Next steps after the calculation
Re-run the calculation with your own figures, stress the two most sensitive inputs by ±10 %, and only then act on the result.
The audience here is people who need one trustworthy number before their next decision. Everything you need to defend the result — formula, substitution, sensitivity, source — is printed above, so the only remaining variable is the honesty of your inputs.