Skip to main content
v2026.11,986 entries · CC-BY 4.0

Range Rule of Thumb: Estimating Standard Deviation from the Range

The range rule of thumb estimates standard deviation as range ÷ 4. See the formula, a checked worked example, its limits, and Excel, R and Python code.

Written and maintained by CASRAI Editorial Board

Published · Last updated

On this page

The range rule of thumb estimates a standard deviation from the range: s ≈ (maximum − minimum) / 4. It is a quick approximation for planning and sanity checks, not a replacement for computing the standard deviation of data you have actually collected.

Ask CASRAI · free to tryQuestion about this guide? Ask CASRAI answers with linked sources. 2 free questions a day, no account.

Ask your first question freeSubscribe

What does the range rule of thumb mean?

The range rule of thumb says that the standard deviation of a data set is roughly one quarter of its range. Introductory texts such as Triola’s Elementary Statistics (14th ed.) use it as: estimated s = Range / 4, where Range = maximum value − minimum value. The idea behind it is that for roughly bell-shaped data most values fall within about two standard deviations of the mean, so the full spread of the data covers about four standard deviations. The rule is also called the range rule, and searches for the “range of thumb rule” and “range rule of thumb statistics” refer to the same method. For a spread measure that uses every value rather than only the two extremes, see mean absolute deviation.

The rule has two uses that are often mixed up: (1) estimating a standard deviation from a range, and (2) working backwards from a mean and standard deviation to the values that count as “usual.” Both are covered below.

Formula

Estimating the standard deviation from the range:

s ≈ Range / 4 = (xmax − xmin) / 4

  • s: the estimated sample standard deviation.
  • xmax: the largest value in the data (or the largest value you expect).
  • xmin: the smallest value in the data (or the smallest value you expect).
  • Range: xmax − xmin.

Going the other way, a standard deviation implies a range of about 4s. Usual values, as the textbook rule defines them, lie within two standard deviations of the mean:

  • Minimum usual value ≈ mean − 2 × s
  • Maximum usual value ≈ mean + 2 × s

The two-standard-deviation boundary comes from the same bell-curve reasoning as the empirical rule; see the guide on the normal distribution and the 68-95-99.7 empirical rule.

Worked example: how to calculate the standard deviation range

Ten measurements: 12, 12, 14, 15, 16, 18, 18, 20, 20, 25.

  1. Range = 25 − 12 = 13.
  2. Range-rule estimate: s ≈ 13 / 4 = 3.25.
  3. Usual values with the estimate: mean = 170 / 10 = 17.0, so 17.0 − 2(3.25) = 10.5 up to 17.0 + 2(3.25) = 23.5. The value 25 falls outside this interval.

Now compute the actual sample standard deviation. The deviations from the mean (17.0) are −5, −5, −3, −2, −1, 1, 1, 3, 3, 8; their squares are 25, 25, 9, 4, 1, 1, 1, 9, 9, 64, which sum to 148. The sample variance is 148 / 9 = 16.44, so the sample standard deviation is √16.44 = 4.06. (The population formula, dividing by 10, gives 3.85.)

The shortcut gave 3.25 against a true 4.06, about 20% too low. With the true value, 17.0 + 2(4.06) = 25.1, and 25 is no longer flagged. The example shows why the rule is an estimate: with only ten values, one extreme observation inflates the range and the quarter-range does not track the real spread. For the exact calculation see how to find sample variance.

Interpretation: how to read the result

Situation How to read the range-rule estimate
Roughly bell-shaped data, n around 30 The setting where the rule is reported to work well (Ramirez & Cox, 2012). Use as a quick planning figure.
Small n (e.g. 10) or a single extreme value Expect a noticeable miss, as in the example above (3.25 vs 4.06). Treat as an order-of-magnitude figure.
Skewed, uniform, exponential or otherwise non-normal data The rule can fail badly; Ramirez & Cox report it “fails miserably” for other distributions and sample sizes.
A value outside mean ± 2s Treated as unusual under the textbook rule. This flags a value for a closer look; it is not proof of an error or an outlier.

Assumptions and common mistakes: what is a normal distribution?

A normal distribution is the symmetric, bell-shaped distribution fully described by its mean and standard deviation; see normal and standard normal distribution. The range rule leans on that shape. Ramirez and Cox (2012), who used Monte Carlo simulation to study the rule, describe it as working well when the data come from a normal distribution and the sample size is around 30, and as failing for other distributions and sample sizes. Common mistakes:

  • Reporting range / 4 as the standard deviation of collected data. When you have the data, compute the actual standard deviation.
  • Applying it to skewed data or data with a few extreme values, where the range is driven by the tails.
  • Treating mean ± 2s as a hard cut-off. It is a rough guide to what is unusual.
  • Confusing the estimate with the standard error. The standard error describes the precision of a mean; see standard error of the mean.

A legitimate use is planning: when you need a standard deviation before collecting data (for example, to size a study), a guess of the plausible minimum and maximum divided by 4 gives a starting figure. See what is a good sample size for a quantitative study for how that figure feeds sample-size thinking, and treat it as a planning input that should be replaced by a real estimate when pilot data exist.

How to report it (APA example)

Report the standard deviation computed from your data, not the range-rule estimate. For the example data: M = 17.00, SD = 4.06, n = 10. If you used the range rule to plan a study, say so and label it as an assumption, for example: “A standard deviation of 3.25 was assumed for planning, estimated as the expected range (13) divided by 4.”

How to calculate it in Excel, R or Python

Excel

=(MAX(A1:A10)-MIN(A1:A10))/4     // range-rule estimate =STDEV.S(A1:A10)                  // actual sample SD

R

x <- c(12,12,14,15,16,18,18,20,20,25) diff(range(x))/4   # 3.25 sd(x)              # 4.055

Python

import statistics as st x = [12,12,14,15,16,18,18,20,20,25] (max(x) - min(x)) / 4   # 3.25 st.stdev(x)             # 4.055 (sample SD)

FAQ

What does the range rule of thumb mean?

It means the standard deviation is about one quarter of the range, s ≈ (max − min) / 4, and that usual values lie within about two standard deviations of the mean.

How do I find the calculated range?

Subtract the smallest value from the largest. In the example, 25 − 12 = 13. To go from a standard deviation back to an approximate range, multiply by 4.

How do I calculate the standard deviation range (usual values)?

Compute the mean, then take mean − 2s and mean + 2s, where s is the actual or estimated standard deviation.

Is there a range rule of thumb calculator?

None is needed: divide the range by 4 on any calculator, or use the Excel, R or Python lines above.

Sources

  • Ramirez, A., & Cox, C. (2012). Improving on the range rule of thumb. Rose-Hulman Undergraduate Mathematics Journal, 13(2). scholar.rose-hulman.edu/rhumj/vol13/iss2/1 (read 2026-11-07).
  • Triola, M. F. Elementary Statistics (14th ed.), Pearson, chapter 3 exercise solutions on estimating the standard deviation with the range rule of thumb (read 2026-11-07).

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · free to try

Ask about Range Rule of Thumb: Estimating Standard Deviation from the Range

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →