Geometric standard deviation¶

A dimensionless multiplicative spread factor obtained by exponentiating the standard deviation of logarithms.

Version: v2 · 2026-09-19 · History
Domain-specific #: 4721
Domain group: Formal Sciences
Origin domain: Experimental Design & Statistics

Core Idea¶

For positive data, GSD equals exp(sd(log X)); in a lognormal model, multiplying or dividing the geometric mean by powers of GSD describes symmetric intervals on the log scale. The logarithm converts ratios to differences, ordinary standard deviation measures log spread, and exponentiation restores a multiplicative factor. The abstraction is therefore identified by a declared carrier, a transformation or constraint over that carrier, and an invariant that tells an analyst whether the named structure is genuinely present.

The load-bearing residual is not the broad topic of statistics. It is the domain-specific identity determined by all observations are positive, the log base and dispersion convention are stated, and the result is interpreted multiplicatively.

How would you explain it like I'm…

Times-or-Divided Spread

Some things grow by 'times' instead of 'plus', like plants that can be twice as tall or half as tall as each other. The geometric standard deviation tells you how many times bigger or smaller things usually are from the middle one. Instead of saying 'give or take 3', it says 'times or divided by 2'.

Multiply-Divide Spread

The usual standard deviation tells how far numbers usually are from the average by adding or subtracting. The geometric standard deviation tells how spread out numbers are by multiplying or dividing. If it is 2, a typical value is about twice or half the geometric mean. It only works when all the numbers are positive, like sizes or amounts. It suits data that grow or shrink by factors.

Multiplicative Spread Factor

The geometric standard deviation measures spread multiplicatively for positive data. You take the logarithm of each value, compute the ordinary standard deviation of those logs, and then exponentiate the result. Because logs turn ratios into differences, this gives a factor rather than an amount, like 'times or divided by 1.5'. The arithmetic standard deviation, by contrast, gives a plus-or-minus amount in the original units. In a lognormal model, the geometric mean multiplied and divided by the GSD gives a range that is symmetric on the log scale. All values must be positive, and the log base and standard-deviation convention should be stated.

For strictly positive observations X, the geometric standard deviation is GSD = exp(sd(log X)), dimensionless and at least 1. The logarithm maps multiplicative variation to additive variation, the ordinary standard deviation measures that log-scale spread, and exponentiation converts it back to a multiplicative factor. In a lognormal model, intervals of the form GM ÷ GSD^k to GM × GSD^k are symmetric on the log scale and correspond to k standard deviations there. Unlike the arithmetic SD, the GSD is interpreted as 'times or divided by' rather than 'plus or minus'. The construction is undefined for zero or negative values. The choice of log base must be consistent with the exponentiation, and whether the SD uses an n or n−1 denominator should be stated. Interpreting the GSD additively is a category error.

Scope of Application¶

Geometric standard deviation belongs to statistics and is useful where the analyst can specify the typed statistics carrier, defining objects and relations, parameters, conventions, evidence, boundary cases and comparison targets, then evaluate all observations are positive, the log base and dispersion convention are stated, and the result is interpreted multiplicatively. The scope is broad within that domain but bounded by the need for all observations are positive, the log base and dispersion convention are stated, and the result is interpreted multiplicatively. The entry records a descriptive analytical identity; practical use requires the governing domain's evidence, standards, and safety obligations.

Clarity¶

The abstraction clarifies a crowded vocabulary by making all observations are positive, the log base and dispersion convention are stated, and the result is interpreted multiplicatively the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test. A bare label is insufficient because the name Geometric standard deviation can be used for a formal identity, an implementation, or a neighboring result unless carrier and convention are stated.

Manages Complexity¶

Without the abstraction, an analyst must reason directly over many local details: the carrier roles, admissibility assumptions, competing conventions, derived invariants, boundary cases, and proof or validation obligations specific to Geometric standard deviation. Geometric standard deviation compresses them into the roles in the structural signature. That compression permits comparison across instances without erasing the variables that determine validity. It also exposes which details may be varied safely and which are constitutive.

Abstract Reasoning¶

Identify the carrier. State what the elements, states, objects, or observations are: the typed statistics carrier, defining objects and relations, parameters, conventions, evidence, boundary cases and comparison targets. Reject examples whose alleged carrier belongs to a different problem.
Lock the constitutive rule. Express all observations are positive, the log base and dispersion convention are stated, and the result is interpreted multiplicatively independently of one notation or implementation.

Knowledge Transfer¶

Knowledge transfers strongly among subfields of statistics because they reuse the typed statistics carrier, defining objects and relations, parameters, conventions, evidence, boundary cases and comparison targets, The logarithm converts ratios to differences, ordinary standard deviation measures log spread, and exponentiation restores a multiplicative factor., and type the carrier, state every parameter and convention in the definition, test that all observations are positive, the log base and dispersion convention are stated, and the result is interpreted multiplicatively, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases.

Relationships to Other Abstractions¶

Current abstraction Geometric standard deviation Domain-specific

Parents (1) — outgoing typed relations

Geometric standard deviation is a kind of Measurement Prime

What makes it its own entry: the domain-specific identity determined by all observations are positive, the log base and dispersion convention are stated, and the result is interpreted multiplicatively.

Hierarchy path (1) — routes to 1 parentless root

Geometric standard deviation → is a kind of Measurement

Neighborhood in Abstraction Space¶

Geometric standard deviation sits in a crowded region of the domain-specific corpus (12^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Descriptive Statistics & Correlation Measures (40 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-25