RSD Calculator

Figran

Paste replicates for every sample to get %RSD or %CV, flags over your limit and a precision figure.

  • Many samples at once
  • Intra- and inter-assay CV
  • Precision figure and methods text
Data
8 samples · 2 replicates
Orientation
Replicates are from
Label
Options
Limit (%)
Standard deviation
Average CV as

Plot
Color
Width
Intra-assay CV 1.70%

Mean of 8 samples · range 0.565–4.04% · 0 above 15% · sample SD (n − 1)

0510152015%Std 0.0488Std 0.04884.04Std 0.195Std 0.1951.73Std 0.391Std 0.3913.02Std 0.781Std 0.7810.565Std 1.56Std 1.560.578Std 3.13Std 3.131.26Std 6.25Std 6.251.57Std 12.5Std 12.50.822CV (%)

Results by sample

Show
NamenMeanSD%CVStatus
Std 0.048820.01750.00074.04%Within limit
Std 0.19520.12250.00211.73%Within limit
Std 0.39120.21050.00643.02%Within limit
Std 0.78120.37550.00210.565%Within limit
Std 1.5620.61150.00350.578%Within limit
Std 3.1321.01000.01271.26%Within limit
Std 6.2521.34900.02121.57%Within limit
Std 12.521.72000.01410.822%Within limit

Summary

Mean of CVs1.70%
RMS2.06%
Median1.42%
Samples8
Range0.565–4.04%
Above limit0 / 8 · 15%
Standard deviationSample (n − 1)

Report

Results

The mean intra-assay CV was 1.70% (range 0.565–4.04%, 8 samples); none exceeded 15%.

Methods

Intra-assay precision was expressed as the %CV of duplicate measurements, calculated as SD / mean × 100 using sample SD (n − 1).

RSD calculator examples

Explore precision calculations using real datasets.

0510152015%Std 0.0488Std 0.04884.04Std 0.195Std 0.1951.73Std 0.391Std 0.3913.02Std 0.781Std 0.7810.565Std 1.56Std 1.560.578Std 3.13Std 3.131.26Std 6.25Std 6.251.57Std 12.5Std 12.50.822CV (%)

ELISA standards in duplicate: intra-assay CV

Eight duplicate standards from one assay run.

Source: R DNase (Davidian & Giltinan 1995) · 8 standards · duplicates

151815201522152415261528Sample 1%RSD 0.206%Peak area

HPLC replicate injections: %RSD

Check the precision of six peak areas.

6 injections · peak area

02040600.0100.101.01015%Std 0.0488Std 0.195Mean ODCV (%)

Inter-assay CV across 11 ELISA runs

Compare plate means for eight standards.

Source: R DNase · 8 standards · 11 runs

010203040500.0100.101.01015%Run 10 · 0.0488 ng/mLRun 4 · 0.0488 ng/mLRun 9 · 0.0488 ng/mLMean ODCV (%)

Duplicate CVs on 11 plates: which wells exceed the limit

Locate high CVs among 88 duplicate pairs.

Source: R DNase · 88 duplicate pairs

0102030400.05010015020025015%Substrate 0.02Mean rateCV (%)

Enzyme rates in duplicate: CV rises at low substrate

Compare replicate enzyme rates over substrate concentrations.

Source: R Puromycin (Treloar 1974)

0510152015%AA3.83BB4.55CC4.09DD4.14EE4.54FF5.42CV (%)

Penicillin potency on 24 plates: inter-plate CV

Compare six samples across plates.

Source: Davies & Goldsmith 1972 (lme4 Penicillin) · 6 samples · 24 plates

0102030Sample 1%RSD 124%Copper (ppm)

Copper in flour: one outlier inflates the RSD

The RSD falls from 123.76% to 21.42% without the largest value.

Source: Analytical Methods Committee 1989 (MASS chem) · 24 determinations

0510152015%High controlHigh control5.07Low controlLow control6.25CV (%)

From mean and SD: saliva cortisol controls

Recalculate CVs for high and low controls.

Source: Salimetrics · 2 controls · 10 plates

How to calculate RSD and CV

Go from replicate measurements to a precision figure in four steps.

  1. 1Paste your data

    Enter one sample per row or paste a single column of values.

  2. 2Choose the run

    Select one run for intra-assay or different runs for inter-assay precision.

  3. 3Review the results

    Check each sample and any values above your limit.

  4. 4Export the figure

    Download the figure or copy the report.

Intra-assay and inter-assay CV

Choose the precision measure that matches how the data were collected.

PrecisionInputCalculationCommon limit
Intra-assay CVReplicates from one runCV for each sample, then average< 10%
Inter-assay CVOne mean per run for each sampleCV for each sample, then average< 15%
Intermediate precisionIntermediate precision accounts for variation across runs under ICH Q2(R2). A variance-components analysis can separate sources of variation; this calculator uses the simpler average of per-sample CVs described by Salimetrics.

Mean or root mean square

For the default duplicate data, the mean CV is 1.70% and the RMS CV is 2.06%. RMS gives greater weight to large CVs.

Precision profile

Plot CV against mean concentration or signal to see how precision changes across the measurement range.

02040600.0100.101.01015%Std 0.0488Std 0.195Mean ODCV (%)

References: Reed et al. (2002) · Salimetrics · ICH Q2(R2) · ICH M10

RSD and coefficient of variation formula

Relative standard deviation and coefficient of variation are synonymous: divide the standard deviation by the mean and multiply by 100.

CV=sx̄×100
s=∑(xi−x̄)2n−1σ=∑(xi−x̄)2n
s=|a−b|2RMS=∑CVi2k

For the six HPLC injections, 3.13603 / 1522.63 × 100 = 0.206%.

References: IUPAC Gold Book · Hyslop & White (2009)

RSD calculator FAQ

Is RSD the same as the coefficient of variation?

Yes. %RSD and %CV both equal SD divided by the mean, multiplied by 100. RSD is common in analytical chemistry, while CV is common in immunoassays.

Should I use the sample or the population standard deviation?

Use sample SD for replicate measurements sampled from possible measurements. Use population SD only when the entire population was measured.

What %CV or %RSD is acceptable?

It depends on the method and its validation requirements. For chromatographic bioanalysis, ICH M10 specifies at most 15% CV, or 20% at the lower limit of quantification; ligand-binding assays have different limits. Set the limit required by your protocol. For ongoing QC monitoring, use the Levey-Jennings Chart Maker.

Should I calculate ELISA CVs from OD or concentration?

Use back-calculated concentration to report sample precision; OD CV describes repeatability of the plate reading. ELISA standard curve calculator.

How do I calculate %CV or %RSD in Excel?

Use =STDEV.S(A2:A7)/AVERAGE(A2:A7)*100. The same formula works in Google Sheets.

Why is the CV so high at low concentrations?

When the mean approaches zero while SD stays similar, CV increases. A precision profile makes that pattern visible. To investigate unusual observations, use the Outlier Calculator.

When should I not use the coefficient of variation?

Avoid CV for quantities without a meaningful true zero or on a logarithmic scale, including Celsius temperature, pH and qPCR Ct values. delta delta Ct calculator.

How many replicates do I need for %RSD?

At least two values are needed to compute sample SD, but the number needed for validation depends on the protocol. ICH Q2(R2) describes repeatability studies using nine determinations or six at the test concentration.

When should I report the CV instead of the standard deviation?

Use CV to compare relative variability across positive means; SD retains the original units. CV does not measure agreement between two methods. Bland–Altman plot.

Check the precision of your replicates

Calculate CVs and create a precision figure from your data.

Calculate %CV