Statistical Symbols and Abbreviations - AMA Manual of Style
Show Summary Details
Page of

# Statistical Symbols and Abbreviations

Chapter:
Study Design and Statistics
Page of

PRINTED FROM AMA MANUAL OF STYLE ONLINE (www.amamanualofstyle.com). © American Medical Association, 2009. All Rights Reserved. Under the terms of the license agreement, an individual user may print out a PDF of a single chapter of a title in AMA Manual of Style Online for personal use (for details see Privacy Policy).

Subscriber: null; date: 27 August 2016

# Statistical Symbols and Abbreviations

The following may be used without expansion except where noted by an asterisk. For a term expanded at first mention, the abbreviation may be placed in parentheses after the expanded term and the abbreviation used thereafter (see also 14.11, Abbreviations, Clinical, Technical, and Other Common Terms). Most terms other than mathematical symbols can also be found in 20.9, Glossary of Statistical Terms.

Symbol or Abbreviation

Description

|x|

absolute value

sum

>

greater than

greater than or equal to

<

less than

less than or equal to

^

hat, used above a parameter to denote an estimate

ANOVA

analysis of variance*

ANCOVA

analysis of covariance*

α

alpha, probability of type I error

1 − α

confidence coefficient

β

beta, probability of type II error; or population regression coefficient

1 − β

power of a statistical test

b

sample regression coefficient

CI

confidence interval*

CV

coefficient of variation (s/x̄) x 100*

D

difference

df

degrees of freedom (v is the international symbol55 and also may be used if familiar to readers)

D2

Mahalanobis distance, distance between the means of 2 groups

δ

delta, change delta

δ

delta, true sampling error

epsilon, true experimental error

e

exponential

E(x)

expected value of the variable x

f

frequency; or a function of, usually followed by an expression in parentheses, eg, f(x)

Fv1,v2(1 − α)

F test, ratio of 2 variances, with df = v1, v2 for numerator and denominator, respectively, and (1 − α) = confidence coefficient

G2(df)

likelihood ratio χ2

H0

null hypothesis

H1

alternate hypothesis; specify whether 1- or 2-sided

κ

kappa statistic

λi

lambda, hazard function for interval i; eigenvalue; or estimate of parameter for log-linear models

Λ

Wilks lambda

ln

natural logarithm

log

logarithm to base 10

MANOVA

multivariate analysis of variance*

μ

population mean

n

size of a subsample

N

total sample size

n!

(n) factorial

OR

odds ratio*

p

statistical probability

$χ32$

χ2 test or statistic, with 3 df shown as an example

r

bivariate coefficient of determination

R

multivariate correlation coefficient

r2

bivariate coefficient of determination

R2

multivariate coefficient of determination

RR

relative risk*

ρ

rho, population correlation coefficient

SD

standard deviation of a difference D

s2

sample variance

σ2

sigma squared, population variance

σ

sigma, population SD

SD

standard deviation of a sample

SE

standard error

SEM

standard error of the mean

t

Student t; specify α level, df, 1-tailed vs 2-tailed

τ

Kendall tau

T2

Hotelling T2 statistic

u

Mann-Whitney U(Wilcoxon) statistic

$x¯$

arithmetic mean

z

z Score