Signicant gures

Leading zeros are not signicant. For example,

0.00052 has two signicant gures: 5 and 2.

The signicant gures of a number are those digits that

carry meaning contributing to its precision. This includes
all digits except:[1]

Trailing zeros in a number containing a decimal

point are signicant. For example, 12.2300 has six
signicant gures: 1, 2, 2, 3, 0 and 0. The number 0.000122300 still has only six signicant gures
(the zeros before the 1 are not signicant). In addition, 120.00 has ve signicant gures since it has
three trailing zeros. This convention claries the
precision of such numbers; for example, if a measurement precise to four decimal places (0.0001) is
given as 12.23 then it might be understood that only
two decimal places of precision are available. Stating the result as 12.2300 makes clear that it is precise
to four decimal places (in this case, six signicant
gures).

All leading zeros;

Trailing zeros when they are merely placeholders to
indicate the scale of the number (exact rules are explained at identifying signicant gures); and
Spurious digits introduced, for example, by calculations carried out to greater precision than that of the
original data, or measurements reported to a greater
precision than the equipment supports.
Signicance arithmetic are approximate rules for roughly
maintaining signicance throughout a computation.
The more sophisticated scientic rules are known as
propagation of uncertainty.

The signicance of trailing zeros in a number not

containing a decimal point can be ambiguous. For
example, it may not always be clear if a number like
1300 is precise to the nearest unit (and just happens
coincidentally to be an exact multiple of a hundred)
or if it is only shown to the nearest hundred due to
rounding or uncertainty. Various conventions exist
to address this issue:

Numbers are often rounded to avoid reporting insignificant gures. For example, it would create false precision to express a measurement as 12.34500 kg (which
has seven signicant gures) if the scales only measured
to the nearest gram and gave a reading of 12.345 kg
(which has ve signicant gures) . Numbers can also
be rounded merely for simplicity rather than to indicate
a given precision of measurement, for example to make
them faster to pronounce in news broadcasts.

A bar may be placed over the last significant gure; any trailing zeros following
this are insignicant. For example, 1300
has three signicant gures (and hence
indicates that the number is precise to the
nearest ten).

Arithmetic precision can also be dened with reference

to a xed number of decimal places (the number of digits
following the decimal point). This second denition is
useful in applications where the number of digits in the
fractional part has particular importance, but it does not
follow the rules of signicance arithmetic.

The last signicant gure of a number

may be underlined; for example, 2000
has two signicant gures.

Identifying signicant gures

A decimal point may be placed after
the number; for example 100. indicates
specically that three signicant gures
are meant.[3]

Specically, the rules for identifying signicant gures

when writing or interpreting numbers are as follows:[2]
All non-zero digits are considered signicant. For
example, 91 has two signicant gures (9 and 1),
while 123.45 has ve signicant gures (1, 2, 3, 4
and 5).

In the combination of a number and a unit

of measurement, the ambiguity can be
avoided by choosing a suitable unit prex. For example, the number of signicant gures in a mass specied as 1300 g
is ambiguous, while in a mass of 13 hg or
1.3 kg it is not.

Zeros appearing anywhere between two non-zero

digits are signicant. Example: 101.1203 has seven
signicant gures: 1, 0, 1, 1, 2, 0 and 3.
1

3 ARITHMETIC
However, these conventions are not universally used, and it is often necessary to determine from context whether such trailing zeros are intended to be signicant. If all else
fails, the level of rounding can be specied
explicitly. The abbreviation s.f. is sometimes used, for example 20 000 to 2 s.f. or
20 000 (2 sf)". Alternatively, the uncertainty
can be stated separately and explicitly with a
plus-minus sign, as in 20 000 1%, so that
signicant-gures rules do not apply. This also
allows specifying a precision in-between powers of ten (or whatever the base power of the
numbering system is).

1.1

Scientic notation

In most cases, the same rules apply to numbers expressed in scientic notation. However, in the normalized form of that notation, placeholder leading and
trailing digits do not occur, so all digits are signicant.
For example, 0.00012 (two signicant gures) becomes
1.2104 , and 0.00122300 (six signicant gures) becomes 1.22300103 . In particular, the potential ambiguity about the signicance of trailing zeros is eliminated.
For example, 1300 to four signicant gures is written as
1.300103 , while 1300 to two signicant gures is written as 1.3103 .

If the rst non-signicant gure is a 5 followed by

other non-zero digits, round up the last signicant
gure (away from zero). For example, 1.2459 as the
result of a calculation or measurement that only allows for 3 signicant gures should be written 1.25.
If the rst non-signicant gure is a 5 not followed
by any other digits or followed only by zeros, rounding requires a tie-breaking rule. For example, to
round 1.25 to 2 signicant gures:
Round half up (also known as 5/4) rounds
up to 1.3. This is the default rounding method
implied in many disciplines if not specied.
Round half to even, which rounds to the nearest even number, rounds down to 1.2 in this
case. The same strategy applied to 1.35 would
instead round up to 1.4.
Replace non-signicant gures in front of the decimal by zeros.
In nancial calculations, a number is often rounded to
a given number of places (for example, to two places
after the decimal separator for many world currencies).
Rounding to a xed number of decimal places in this way
is an orthographic convention that does not maintain signicance, and may either lose information or create false
precision.

The part of the representation that contains the signicant As an illustration, the decimal quantity 12.345 can be exgures (as opposed to the base or the exponent) is known pressed with various numbers of signicant digits or decimal places. If insucient precision is available then the
as the signicand or mantissa.
number is rounded in some manner to t the available
Alternatively: 1. All non-zero digits are signicant 2. In a
precision. The following table shows the results for varnumber without a decimal point, only zeros between nonious total precisions and decimal places rounded to the
zero digits are signicant. 3. In a number with a decimal
nearest value using the round-to-even method.
point, all zeros to the right of the rst non-zero digits are
The representation of a positive number x to a precision
signicant.
of p signicant digits has a numerical value that is given
by the formula:

Rounding and decimal places

The basic concept of signicant gures is often used in

connection with rounding. Rounding to signicant gures is a more general-purpose technique than rounding
to n decimal places, since it handles numbers of dierent
scales in a uniform way. For example, the population of
a city might only be known to the nearest thousand and
be stated as 52,000, while the population of a country
might only be known to the nearest million and be stated
as 52,000,000. The former might be in error by hundreds,
and the latter might be in error by hundreds of thousands,
but both have two signicant gures (5 and 2). This reects the fact that the signicance of the error (its likely
size relative to the size of the quantity being measured) is
the same in both cases.
To round to n signicant gures:[4]

round(10n x)10n , where n = oor(log10 x) +

1 p.
For negative numbers, the formula can be used on the
absolute value; for zero, no transformation is necessary.
Note that the result may need to be written with one of
the above conventions explained in the section Identifying signicant gures to indicate the actual number of
signicant digits if the result includes for example trailing
signicant zeros.

3 Arithmetic
Main article: Signicance arithmetic

3
A common guide often used when performing calcula- 0.0031 == 1.004.[6]
tions by hand is as follows.
For multiplication and division, the result should have as
many signicant gures as the measured number with the
smallest number of signicant gures.

6 Estimation

For addition and subtraction, the result should have as Main article: Estimation
many decimal places as the measured number with the
smallest number of decimal places (for example, 100.0 + When estimating the proportion of individuals carrying
1.111 = 101.1).
some particular characteristic in a population, from a ranIn a base 10 logarithm of a normalized number, the result dom sample of that population, the number of signicant
should be rounded to the number of signicant gures in gures should not exceed the maximum precision allowed
the normalized number. For example, log10 (3.000104 ) by that sample size. The correct number of signicant
= log10 (104 ) + log10 (3.000) 4 + 0.47712125472, gures is given by the order of magnitude of sample size.
This can be found by taking the base 10 logarithm of samshould be rounded to 4.4771.
ple size and rounding to the nearest integer.
When taking antilogarithms, the resulting number should
have as many signicant gures as the mantissa in the For example, in a poll of 120 randomly chosen viewers of a regularly visited web page we nd that 10 peologarithm.
ple disagree with a proposition on that web page. The
When performing a calculation, do not follow these
order of magnitude of our sample size is Log10 (120) =
guidelines for intermediate results; keep as many digits
2.0791812460... which rounds to 2. Our estimated proas is practical (at least 1 more than implied by the preportion of people who disagree with the proposition is
cision of the nal result) until the end of calculation to
therefore 0.083, or 8.3%, with 2 signicant gures. This
[5]
avoid cumulative rounding errors.
is because in dierent samples of 120 people from this
population, our estimate would vary in units of 1/120,
and any additional gures would misrepresent the size of
4 Estimating tenths
our sample by giving spurious precision. To interpret our
estimate of the number of viewers who disagree with the
When using a ruler, initially use the smallest mark as the proposition we should then calculate some measure of our
rst estimated digit. For example, if a rulers smallest condence in this estimate.
mark is cm, and 4.5 cm is read, it is 4.5 (0.1 cm) or 4.4
4.6 cm.
It is possible that the overall length of a ruler may not be 7 Relationship to accuracy and preaccurate to the degree of the smallest mark and the marks
cision in measurement
may be imperfectly spaced within each unit. However
assuming a normal good quality ruler, it should be possible to estimate tenths between the nearest two marks to Main article: Accuracy and precision
achieve an extra decimal place of accuracy. Failing to do
this adds the error in reading the ruler to any error in the In various technical elds, accuracy refers to the closecalibration of the ruler.
ness of a given measurement to its true value; precision
refers to the stability of that measurement when repeated
many times. The number of signicant gures roughly
corresponds to precision, not accuracy.

Calculation

When multiplying several quantities, the number of signicant gures in the nal answer is the number of signicant gures in the factor having the least number of
signicant gures (the least precise). For example, given
the two measurements 16.3 cm (0.1 cm) and 4.5 cm
(0.1 cm), the range is: 16.2*4.4 16.4*4.6 (71.28
75.44) cm2 and the average is 73.36; however, only two
signicant gures (i.e. 73 cm2 ) can be claimed in the
result (the area calculated).

8 In computing
Main article: Floating point

Computer representations of oating point numbers typically use a form of rounding to signicant gures, but
with binary numbers. The number of correct signicant
gures is closely related to the notion of relative error
When adding numbers, the number of decimal places in (which has the advantage of being a more accurate meathe result is the smallest of the number decimal places in sure of precision, and is independent of the radix of the
any term. For example, 123 + 5.35 == 128 and 1.001 + number system used).

12

See also
Accuracy and precision
Benfords Law (First Digit Law)
Engineering notation
Error bar
False precision
IEEE754 (IEEE oating point standard)
Interval arithmetic
Kahan summation algorithm
Precision (computer science)
Round-o error

10

References

[1] Chemistry in the Community; Kendall-Hunt:Dubuque, IA

1988
[2] Giving a precise denition for the number of correct signicant digits is surprisingly subtle, see Higham, Nicholas
(2002). Accuracy and Stability of Numerical Algorithms
(2nd ed.). SIAM. p. 3.
[3] Myers, R. Thomas; Oldham, Keith B.; Tocci, Salvatore
(2000). Chemistry. Austin, Texas: Holt Rinehart Winston. p. 59. ISBN 0-03-052002-9.
[4] Numerical Mathematics and Computing, by Cheney and
Kincaid.
[5] de Oliveira Sannibale, Virgnio (2001). Measurements
and Signicant Figures (Draft)" (PDF). Freshman Physics
Laboratory. California Institute of Technology, Physics
Mathematics And Astronomy Division. Archived from
the original on June 18, 2013.
[6] Serway, Raymond A. (1990). Physics for Scientists
and Engineers: With Modern Physics. Saunders Golden
Sunburst Series (3rd ed.). Harcourt School. ISBN
0030313538.

11

Further reading

Subcommittee E11.30 (2013). Standard Practice

for Using Signicant Digits in Test Data to Determine
Conformance with Specications. ASTM International. doi:10.1520/E0029-08. ASTM E29-13.

12

External links

Signicant Figures Calculator

Signicant Figures Video by Khan academy

EXTERNAL LINKS

13
13.1

Text and image sources, contributors, and licenses

Text

Signicant gures Source: http://en.wikipedia.org/wiki/Significant%20figures?oldid=648469892 Contributors: Patrick, D, Michael

Hardy, Dominus, Tango, Georey, Jasper, AugPi, SaveThePoint, Alex S, Davidzuccaro, Jitse Niesen, Markhurd, Maximus Rex, Omegatron, Indefatigable, Pakaran, Fredrik, Nurg, Jxg, Hadal, HaeB, SpellBott, Giftlite, Djinn112, Jao, BenFrantzDale, 0x6D667061, Bensaccount, Jackol, Bobblewik, Teria, Noe, Antandrus, Beland, Na89, Thincat, Starx, Neutrality, Quota, Irrelevant, Discospinster, Rich Farmbrough, Cfailde, Vsmith, Notinasnaid, Calair, Triona, Bobo192, AnyFile, AdamX, A.b.haydar, La goutte de pluie, Zetawoof, Nsaa, Jumbuck, Musiphil, Velella, TenOfAllTrades, Woodstone, MIT Trekkie, Oleg Alexandrov, Stemonitis, Mel Etitis, Georgia guy, David Haslam,
Robert K S, WadeSimMiser, NebY, Pjetter, Agorf, Cat5nap, NeonMerlin, Platypus222, Yamamoto Ichiro, FlaBot, Winterstein, Mathbot,
AJR, Kolbasz, Fresheneesz, TeaDrinker, Glenn L, Chobot, Benlisquare, Wavelength, Jimp, Michael Slone, Geologician, Chris Capoccia,
Chaser, Manop, Doctor Whom, Calumny, Seeaxid, Leotohill, Goodeye, Poochy, Open2universe, Ninly, Imaninjapirate, Closedmouth,
Pb30, GraemeL, Mais oui!, NeilN, John Broughton, DVD R W, Kurtisnelson, SmackBot, Incnis Mrsi, Slashme, InverseHypercube, Hr2,
Clpo13, Pkirlin, Delldot, Eskimbot, Imzadi1979, Zephyris, Gilliam, Chris the speller, Qwasty, CyberSach, Octahedron80, Patdw, Zhinz,
MrC539, Metallurgist, Alphathon, Sephiroth BCR, Vanis314, UU, Cybercobra, Shadow1, ArglebargleIV, Ong elvin, JoshuaZ, JorisvS,
Mgiganteus1, Beefball, Michael Greiner, Peter Horn, Hu12, Iridescent, Twas Now, GDallimore, A. Pichler, Lindsey8417, Tawkerbot2,
Eastlaw, CmdrObot, NSD Student, CBM, Jsd, Djsonik, Outside Center, Spookpadda, Epbr123, N5iln, Andyjsmith, Mojo Hand, Pjvpjv,
Ufwuct, RFerreira, Mentisto, Sidasta, WinBot, Jq4bagiq, Epischedda, Jj137, Salgueiro, Lklundin, Tomertomer, LittleOldMe, .anacondabot, Connormah, VoABot II, Avicennasis, Indon, KOMPRESSOR, DerHexer, Philg88, Stephenchou0722, Epylar, Hdt83, MartinBot,
R'n'B, J.delanoy, PCock, Leon math, Trusilver, Rlsheehan, Washacked3, Ginsengbomb, Cephyr, MikeEagling, Plasticup, KylieTastic,
WarFox, TWCarlson, Goyston, Rucha58, Danny252, Seattle Skier, Philip Trueman, TXiKiBoT, Technopat, SueHay, Faulkner891, Adiamas, Jeremy Reeder, Cuddlyable3, GiZiBoNG, Bstar1, Wolfrock, Logan, The Random Editor, Scarian, Yintan, Redhookesb, Android
Mouse, Tiptoety, Oxymoron83, Antonio Lopez, Johnanth, D14C050, Joelster, ClueBot, The Thing That Should Not Be, Bored0000000,
Roibeird, Excirial, Three-quarter-ten, Brianbjparker, Horizonwards, Mczack26, LC Dave, Johnuniq, SoxBot III, Tompana82, Erodium,
AlanM1, Hotdog20088, Ymc331, Supermanwonderbrain, Weitzhandler, WikiDao, Addbot, Fgnievinski, Neodop, Jncraton, Fluernutter, Ehrenkater, The Reader who Writes, PV=nRT, Fraggle81, Les boys, Sirsparksalot, Georgexu316, AnomieBOT, TipoApuesto333,
Myusername6789, Citation bot, Haverkamp, Gsmgm, Znex, Iaincmac, Cureden, Tmkkeller, JimVC3, Ultimateria, RibotBOT, Amaury,
Smallman12q, Alain tesio, Edgars2007, Magic.Wiki, FrescoBot, Fortdj33, Nickswalker, Kwiki, Pinethicket, I dream of horses, Johnnylake2010, GoodwillTKE, BRUTE, Jschnur, RedBot, SectorSweep, Tofutwitch11, Vrenator, Adi4094, CuboneKing, DARTH SIDIOUS 2,
Regancy42, NerdyScienceDude, Acather96, TheWilsn, Wikipelli, Ida Shaw, F, Hazard-SJ, Inniverse, Wayne Slam, Tolly4bolly, Matthagy,
Sven nestle2, L Kensington, Donner60, Radar999, Jessdarcy, DASHBotAV, Valorousdeath, Xanchester, ClueBot NG, Jnorton7558, MelbourneStar, This lousy T-shirt, Pizza1016, Bped1985, KamiCrit, Tideat, Frietjes, Oghin, O.Koslowski, Karlbonner1982, Widr, Helpful
Pixie Bot, BG19bot, Bmusician, Domlightfoot, Tallytep, AmieKim, D Namtar, Snow Blizzard, Nathanjbrown, Glacialfox, FuzzyCatPotato, Uopchem2519, Simeondahl, 14jbella, Rinkle gorge, Frosty, Brirush, The Anonymouse, Faizan, Melonkelon, Wamiq, Ginsuloft, I
b, Lightningcougar, TheQ Editor, Navstar55, Thunt99612, Jalisco198, GaminCrazy and Anonymous: 662

13.2

Images

File:Question_book-new.svg Source: http://upload.wikimedia.org/wikipedia/en/9/99/Question_book-new.svg License: Cc-by-sa-3.0

Contributors:
Created from scratch in Adobe Illustrator. Based on Image:Question book.png created by User:Equazcion Original artist:
Tkgd2007

13.3

Content license

Creative Commons Attribution-Share Alike 3.0