An Extension of Shapiro and Wilk's W Test for Normality to Large Samples

Author(s): J. P. Royston
Source: Journal of the Royal Statistical Society. Series C (Applied Statistics), Vol. 31, No. 2
(1982), pp. 115-124
Published by: Blackwell Publishing for the Royal Statistical Society
Stable URL: http://www.jstor.org/stable/2347973
Accessed: 20/09/2010 03:57

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=black.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Royal Statistical Society and Blackwell Publishing are collaborating with JSTOR to digitize, preserve and
extend access to Journal of the Royal Statistical Society. Series C (Applied Statistics).

http://www.jstor.org
Appl.Statist.(1982),
31, No. 2, pp. 115-124

An Extensionof Shapiro and Wilk's W Test for

Normalityto Large Samples
By J. P. ROYSTON
MRC ClinicalResearchCentre,WatfordRoad, HarrowHA] 3UJ, Middx.,UK
[ReceivedFebruary1981.RevisedDecember 1981]

SUMMARY
Shapiroand Wilk's(1965) W statisticarguablyprovidesthe best omnibustestof
butis currently
normality, to samplesizesbetween
limited 3 and50.Wisextended up to
n = 2000 and an approximate normalizing transformation suitablefor computer
implementationis given.A novelapplication
ofWintransforming datato normality
is
usingthethree-parameter
suggested, lognormalas an example.
Keywords: TESTSOF NORMALITY;OMNIBUSTEST;W STATISTIC;NORMALIZING
TRANSFORMATION;
POLYNOMIALSMOOTHING

1. INTRODUCTION
RESEARCHinto tests of non-normalitywas given new impetus with the introduction of the so-
calledanalysisofvariancetestbyShapiroand Wilk(1965).The teststatisticWwasconstructed
by consideringthe regressionof orderedsample values on correspondingexpectednormal
orderstatistics,
whichfora samplefroma normallydistributedpopulationis linear;Wwas
obtained as an F-ratio from generalized least-squares analysis to judge the adequacy of the
linear fit.Percentage points of the null distributionof Wwere tabulated forp = 0 01, 002, 005,
0 1, 0 5, 0-9,0 95, 0 98 and 0 99, forsamplesizesn = 3 (1) 50. A normalizingtransformation
for
Wusinga JohnsonSB approximationin the regionn = 7(1)50 was laterproposed(Shapiro
and Wilk, 1968), though tables were still required for n = 4(1) 6.
Extensiveempiricalcomparisonsof Wwithothertestsof non-normality
usingcomputer-
generated pseudo-random numbers indicated that W had good power properties against a
wide range of alternativedistributions,and was thereforetrulyan omnibus test (Shapiro et al.,
1968).Subsequently, otherstatisticsofthe Wtype,namelyY(D'Agostino, 1971),W' (Shapiro
and Francia,1972)and r (Filliben,1975),weredevelopedand shownto have powerproperties
broadlycomparablewiththose of W Anotherapproach was to combine ,lb and b2, the
standardsample measuresof skewnessand kurtosis,into an omnibustest(D'Agostino and
Pearson, 1973; Bowman and Shenton,1975).One such statisticis K2 (Pearson et al., 1977).
The power characteristics of the omnibus tests may be summarizedas follows.For
symmetricalplatykurtic distributions(l/,B= 0, 32 < 3), and formostskewdistributions, Wis
optimal(Pearsonet al., 1977),and is sometimesmarkedlymorepowerfulthanitsnearestrival.
K2, W' and r are all more powerfulthan Wforsymmetric, leptokurticpopulations(B2> 3),
though Wseems littleinferiorforverylong-tailed alternatives,e.g. fB2 = 11. W' and r appear to
have very similar power (Filliben, 1975).
With the exception of Wfor n = 7 (1) 50, the existingomnibus test statisticssufferfromthe
serious disadvantagethat theirsignificancelevels have to be obtained fromtables, and
thereforethe testsare inconvenientforcomputerimplementation. FurthermoreW, whichit
maybe arguedformsthebestgeneraltest,cannotbe calculatedforsampleslargerthan50. The
purpose of the presentpaper is to extend W to sample sizes n -2000, and to develop a
transformation of thenull distribution
of Wto approximatenormalitythroughouttherange
7? n?2000. A methodforcalculatingthesignificance levelof Wforn<7 is also derived.All
( 1982 Royal StatisticalSociety 0035-9254/82/31115 $2.00
116 APPLIED STATISTICS

thetechniquesare easilyprogrammable.Finally,a novelapplicationofusingWtofitthethree-

parameterlognormaldistributionis described.

2. CALCULATION OF W
Let m' = (i1,..., m) denote the vector of expected values of standard normal order
statistics,
and let V= (vii)be the correspondingn x n covariancematrix;thatis

E(xi) = mi(i = 1,...,n) and cov (xi,xj) = vij(i,j = 1,...,In),

wherex1 <x2 <... <xn is an orderedrandom sample froma standardnormaldistribution
N(O,1). Suppose y' = (Y1, ..., Yn) is a randomsampleon whichthe Wtestofnormalityis to be
carriedout, orderedY(1) < Y(2) <*.. < Y(n). Then
n - 2 /n
HZ aiy() /(iY)
where
a' = (aa,...,) = m'V-1[(m'V-1)(V- m)]-2

The coefficients
{ai} are the normalized"best linearunbiased"coefficients
tabulatedfor
n?20 by Sarhan and Greenberg(1956). The covariancematrixVwhichfeaturesin a may be
obtained using the -algorithmof Davis and Stephens(1978). However, V is not required
explicitly,and Shapiro and Wilk (1965) offera satisfactoryapproximationfor a which

TABLE 1
Sometheoretical
moments
(pi) and simulation
moments(/i)of W Valuesin thepresentstudyare
based on 6000 simulations
per samplesize

Al 1,/2 ,Xbj b2

Shapiro This Shapiro This Shapiro This Shapiro This

n ji and Wilk study and Wilk study and Wilk study and Wilk study

7 09123 09120 09113 00547 00542 -132 -1 23 641 544

8 0-9174 09175 09174 00497 00499 -1 38 -127 711 544
9 0-9221 09215 09227 00479 0-0466 -1 60 -148 845 6-49
10 09264 09260 09277 00444 00432 - 167 - 1-47 9 28 6-45
15 09424 09422 09423 00320 00323 -1 89 -1 52 1674 647
20 09523 09527 09523 00255 00253 -228 -155 3259 776
30 0-9626 0-9631 00185 00183 -2-73 -1 46 71 77 673
40 09682 09684 00151 00147 -317 -134 13648 617
50 09714 09718 00124 00126 -332 -1 20 21243 605
60 09738 00110 -104 508
70 09755 00099 -1 02 5-64
80 09768 00090 - -092 502
90 09778 00083 -084 4-17
100 09786 00078 -073 386
125 09799 0-0069 -058 3-48
150 09810 00063 -056 369
200 09821 00054 -044 360
300 09835 00044 -0 25 318
400 0-9844 0-0037 -016 3-17
500 09850 00033 -0 10 3-00
750 0-9860 00026 -0-02 3-06
1000 09866 00022 0 05 3-16
1500 0-9876 0 0018 0-18 309
2000 09884 00015 0 24 - 3-16
 TESTING NORMALITY OF LARGE SAMPLES 117

a has the property

improveswithincreasingsample size n, whichwe adopt. By definition,
a'a = 1. Let a* = m'V-'; approximationsa^ fora* are

02mi, i = 23,
... ,n-1,
=t( z nE,
a--
a* , na-*2),i=1 n, . (2.1)

where
2 ^2 g(n-1), n< 20,
a,= an n>20
)g(n),
and
F(2{n + 1])
g(n) - 12f( n+ 1)

Using Stirling'sformula,g(n)may be approximatedand simplified:

n
g(n) [6n + exp(1)Ln+ )2 (2.2)

For n = 6,theexactvalueofg(n)is 0 39166comparedwithan approximatevalue givenby(2.2)

of039164. We have used theapproximations therange7? n?2000,
(2.1) and (2.2) throughout
but exact values of the {aj} forn<7.

Expected normal order statistics

The values of themiforsamplesizes up to 400 weregivenby Harter(1961). To facilitate
computercalculationsof W,however,we constructed a newapproximationformiin therange
2? n?2000, accurateto 0 0001 (Royston,1982).The algorithm,
based on Blom's(1958,pp. 69-
71) formula,has theadvantageofrapiditywhencomparedwith"exact"calculationsinvolving
numericalintegration or summationof series.

3. EXTENSION OF W
Wwas calculatedusingtheformulae(2.1) and (2.2).Values of Wweregeneratedfromsetsof
normal deviates,calculated as the normal integraltransformof uniformpseudo-random
numbers.Wichmannand Hill's (1982) random numbergeneratorwas used. Six thousand
values of W foreach sample size n = 7 (1) 30 (5) 100, 125, 150,200(100) 600, 750, 1000, 1250,
1500 and 2000 weresimulated.Different setsof pseudo-randomnumberswereused foreach
simulation,to avoid dependencebetweenresults.Selectedempiricalfirstand secondmoments
of W are givenin Table 1, and may be comparedwithShapiro and Wilk's (1965, Table 4)
values,whichare exactforE(W) at n< 20 but otherwisebased on extensivesimulation.Use of
theapproximation(2.1) forthedefining coefficients {aj} has had littleeffect
on themomentsof
W1evenat n = 7. A curiousfeatureofTable 1 is thestrongdisagreement betweenour estimates
of 1fB and f2and thosegivenby Shapiro and Wilk. Accordingto our data, skewnessand
kurtosisboth reducein magnitudeas n increasesfrom10, reachingapproximatelynormal
valuesat about n = 750 and thendiverging. Shapiroand Wilkgive>/B = -3 32,f2= 212 43
at n = 50, and stillrisingsharply.We feelthattheirvalues mustbe wrong,and perhapsthat
roundingerrorspoiledtheircalculations.A glanceat theempiricalcumulativedistribution of
W(Fig. 1,and Shapiroand Wilk,Fig. 4) confirms thatWisbecominglessskewas n goes from5
to 50.
118 APPLIED STATISTICS

W 1.00-
1000005

0.96-
82o

0.88-

0.84-
0.92 -
0.80-

1 2 5 10 20 30 50 70 80 90 95
Percentile
plotoftheempiricalcumulativedistribution
FIG. 1.Normalprobability ofW withsamplesizes7, 8, 10,20,50,
function
70, 100,200, 1000.

4. APPROXIMATIONS FOR SIGNIFICANCE LEVEL OF W

Shapiro and Wilk (1965) used Johnson's(1949) SB systemto approximatethe null
of W1viz.
distribution
z= + log{(W- 0/0 -W)1,
wherez is a standardnormaldeviateand y,6, e are constantsforeach n. This transformation
providedan apparentlyadequate fitforthe percentagepointsof Win the rangen = 7(1)50.
The distributionof W is not asymptotically normal (De Wet and Venter,1972), but our
simulationresultsindicatethat the distributionappears to pass quite close to normality
around n = 750 (Table 1). To reflectthisbehaviour,we selectedthe followingapproximate
normalizingtransformation forW:
y = (1- W)' and z = (y- y)a/y (4.1)

wherez is a standardnormaldeviate,and pvand ovare themean and s.d. ofy. Equation (4.1)
allows one to calculatethesignificance leveloftheoriginalW Large valuesofz (as opposed to
small values of W) indicatedeparturefromnormality.The quantitiesi., ,y and ay are all
functionsof n.
Estimatesof A wereobtainedas follows.For each n, the correlationcoefficient between
selectedempiricalquantilesof(1 - W)A(the0 1,0 5, 1,2, 3, 5, 10,20, 30,40, 50,60, 70,80,90, 95
and 99 percentagepoints)and theirnormalequivalentdeviateswas maximizedwithrespectto
2 using simplelinear search.Differential weightingaccordingto the varianceof a normal
quantilewas applied.Since thedefinition ofthedefiningcoefficients {la} changesat n = 20, a
"kink"in thefunctioni. = O(n)was anticipated;therefore thenumberofsimulatedvalues of W
used to estimateA was doubledto 12000 forn = 7 (1) 14,trebledforn = 15(1) 17 and 21 (1) 23,
and quadrupledforn = 18(1)20. A slightminimumforA appears at n = 17 and a sudden
downward step at n = 21 (Fig. 2). The estimatesof 2 were smoothed using separate
polynomialsin log(n)-d, whered = 3 forn = 7-20 and d = 5 forn = 21-2000 (Table 2).
The parametersofthedistribution oftransformed W,[y and aywereestimatedas themean
and s.d. of the y's, recalculatedusing smoothedvalues of 2.(Fig. 3). Their logarithmswere
smoothedwithpolynomialsin log (n)- d, again in two parts,n< 20 and n>20 (Table 2).
 TABLE 2
for calculationof X, loge(f y) and 1oge (Uy) for the normalizin
Polynomialcoefficients
y = (1 - W)'. Values ofz = (y - iy)/uyhave standard normaldistributioniftheoriginalsam
was calculatedis normal;large values of z indicatenon-normality. The polynomialsar
Eci(log n-d)i, whered = 3 for n< 20, d = 5 for 21 < n< 2000

Coefficients

Parameter Range ofn 0 1 2 3 4 5

7-20 0-118898 0-133414 0-327907

21-2000 0480385 0-318828 0 -0 0241665 0-00879701 0-002989
I ogeu
(y) 7-20 -037542 -0-492145 -1124332 -0 199422
21-2000 -191487 - 137888 -0-04183209 0-1066339 -0-03513666 -001504614
log(CY) 7-20 -315805 0-729399 3-01855 1-558776
21-2000 -3-73538 - 1-015807 -0331885 0-1773538 -0-01638782 -0-0321501

TABLE 3
for thecalculationof U3 fromUn, n = 4,5,6. U3 may thenbe convertedto
Coefficients
nificancelevelforW (see text)

Sample
size, n Range ofun aO a, a2 a3 a4

4 -38 to 14t -126233 1-87969 0-0649583 -0-0475604 -0-0139

14 to 86+ -0287696 1-78953 -0-180114 0 0
5 -3-0 to 1-4t -228135 2-26186 0 0 -0-0086
14 to 58t - 163638 560924 - 3-63738 1-08439 0
6 -10 to 1-4t -3-30623 2-76287 -0-83484 1-20857 -050759
1-4 to 541 -5991908 21-04575 -2458061 13-78661 -2-8352

tt U3 = aO + aj un+ a2 u?+ a3 u+n a4 u4.

n

exp(a0?al x?a2 x 2 +a3x 3+a. a x4); X =

U3 =x(+aX+2 3 3
U3= 109ge(Un).
120 APPLIED STATISTICS

A 1.6-

1.4-

1.2-

1.0

0.8-

0.6-

0.4-

0.2-

0 l II I I IIIIIII 1I
I I111 I I I111 I1
7 10 20 30 50 100 200300 500 1000 2000
n
FIG.2. Power,A,requiredin thetransformation of W to normality, ofsamplesize.y = (1 - W)A.Abscissa
as a function
is sample size (log scale). Solid line representssmoothingpolynomials,witha break at n = 20.

Operationalexample
Suppose a sample of size n = 137 yields W= 09620. Referringto Table 2,
x = log(137)-5 = -0080. Then
A = 0 480385+(0 318828)(-0 080) + ... +(0 002989646)(-0 080)5
= 0 4549,
log uy= -1P91487+(- 1 37888)(-0-080)+...+(- 001504614)(-0 080)5
= - 1 8049,
Iuy= 0-1645,
loguy= -3-73538 +(- 1-015807)(-0 080)+...+(0 003852646)(0080)6
= - 3 6563,
uy= 0 02583.
So
z = (y -y)/Uy
= [(1 -09620) 4549-01645]/0-02583
=238 (p<001).
non-normalat the 1 per centlevel.
The sample would be judged significantly

of thenormalizing
Effectiveness transformation
Adequate transformations of W were achievedfornk 7, as judged by the skewnessand
and leptokurtic;
left-skewed
ofy are slightly
kurtosisofy (Fig. 4). The distributions overthe50
sample sizes, the mean (s.d.) of ,lb was -0012 (0032), and of b2 was 3081 (0066).
Approximate95 per centrangesfor,lb, and b2 in normalsamplesof size 6000 are 0+0 063
and 3 + 0 126 respectively.

The smallestsamplesizes
the power transformation
Unfortunately, of W was consideredinadequate for3 < n1< 7.
Wilkand Shapiro(1968) tabulatedselectedvalues ofti = log {(W- e)/(l - W)} againstWand
thestandardnormaldistribution, 1)
wheree, a2 n/(n- is theminimumattainablevalue of W
 TESTING NORMALITY OF LARGE SAMPLES 121

My 1 -0.1 ry

0.5 0.05

0.1 0.01
0.05 0.005

0.01- 0.001
0.005 0.0005

0.001 , 0.0001
7 10 20 30 50 100 200 50010002000
n
FIG. 3. Mean (solid line) and s.d. (brokenline) of y = (1 - W)K.Values are forsmoothedvalues of A.All scales are
logarithmic.
v b, 0.08-
0.06-
0.04 - . 0

0.02- *v..@
.
0 . *
-0.02 - . *
-0.04 . * . .

-01081 :1 1
b2 3.2 - . *X

3.1 -. *. *.* * *X X

3.0 - *

2.9 i 11l I l1 11111

7 10 20 30 50 100 200300 500 1000 2000
n
FIG 4. Skewnessand kurtosisof y. Each pointrepresents
6000 simulatedvalues ofy.

forsamplesize n (Shapiroand Wilk,1965).For n = 3 theexactprobabilitydensityfunctionof

Wis known(Shapiro and Wilk, 1965,p. 595) to be
f(W) = (3/XK)(1
- W) 2
W , 4 WS 1 (4.2)
The corresponding
cumulativedistribution
functionis
(sin-'l1W-sin
p = F(W) = (6/mr) 1 ). (4.3)
In termsof u, withE = 3
forn = 3, one obtains

p {sin (x2(u)
I sinn 4} (4.4)

Wilk and Shapiro (1968, Table 1) give values of un(n = 3,4,5,6) and corresponding
standard normal quantiles Z,, obtained by simulation.Given Zn, one can calculate an
equivalent value of 1u3 via the standard normal integral, W3 ((4.3) inverted) and
122 APPLIED STATISTICS

= log
[(W3 - )/(1 - W3)]. This providessets ofpairedvalues(Un,U3) fora rangeofp values
at each sample size n = 4,5,6.
One thenfindssuitablefunctions to predictU3 fromu,. Givenan observedWwhosep value
is required,the sequence of operationsis
W -+ Un-+ U3 -+ p.
The last stage is accomplishedby (4.4).
to
Polynomialswereused to fitU3 to un (Table 3). The rangeof applicationis restricted
approximately 0 0002< p < 0 9998.
As an example,suppose n = 5, W= 0 690. For n = 5, E = 0 5523,so
0 690-0 5523
U5= log -8115,
0
1-0 690
U3 = -2 28135+226186(-0 8115)+(-000865763)(-0 8115)4
= -4 121,

-{sin1 I
p =p 7E{i -sin - V
|1 + exp(- 4 )24)s-
12A/4}

= 0 00883.
This resultis close to the value p = 0-0088 obtained by linear interpolationin Wilk and
Shapiro'sTable 1.
5. APPLICATION
Wmaybe put to workin thefieldofestimationand
Besidesitsrole as a testofnormality,
data transformation. one wishes to transformdata to (approximate)
Quite frequently,
and in so doingvaluesforseveralparametersmayhave to be found.For example,
normality,
theJohnsonSB transformation, whichaims to generatea standardnormaldeviatez using
z = y+5y,
where
y = log [(x - 0)/(4+ A - x)]
necessitatesestimationof thefourparametersy, 5,4 and A. Severalmethods(e.g. maximum
likelihood,methodofmoments)have beendevisedto do this.An alternative way to maximize
the normalityof thevariabley is to findvalues of 4 and A whichmaximizeWforthey's. In
general,thisprocedurehas theadvantagesofreducingbytwothenumberofparametersto be

TABLE 4
Values ofc in log(weight-c) as estimatedby themethodofmaximumW(MMW).
Data fromRona and Altman(1977)

Raw data data

Transformed Estimatesofc
Age Number
(years) ofboys l/b1 b2 lb1 b2 W MMW Rona and Altman

5 570 091 4-83 -001 329 09866 9.1 8

6 606 059 386 001 3 19 09851 52 8
7 659 1-06 5-28 -0 06 3-68 0 9900 12 2 12
8 612 090 459 002 348 09844 109 12
9 616 1 60 7-55 -0 05 3 99 0-9850 15 8 16
10 611 1 44 5 58 -0 04 4 19 0 9834 17 8 16
11 309 1 39 6 09 0-02 3-83 0 9777 15 2 16
 TESTING NORMALITY OF LARGE SAMPLES 123

and ofprovidingan immediatemeasureof

estimated(sinceWis scale and originindependent),
thesuccessof the normalizingtransformation (via thep value forW). As an illustration,
we
now apply this "methodof maximumW" (MMW) to data forwhichthe three-parameter
lognormaldistribution was an adequate fit.
In a studyof children'sheight,weightand tricepsskinfold,Rona and Altman(1977)
calculatedconstantsc such that
log (weight- c)
was reasonablynormallydistributed. A different
value ofc was obtainedforeach one-yearage
groupin therange5-12,forboys and girlsseparately.Using theraw data (bycourtesyofthe
authors),we givetheestimatedvaluesofc forboysaged 5-12 accordingto theMM WinTable
4. The values of c were foundby simpleone-dimensionalsearch.For all age groups,the
residualskewnessis verysmall,but some leptokurtosis (b2> 3) remains(not platykurtosis,
as
Rona and Altmanincorrectly state).The valuesofc agreewellwiththeroundedvalueschosen
by Rona and Altman,whoseestimateswerea subjectivecompromisebased on ,lb and b2for
the transformed data.
The Wmethodis easy to apply and, in thisexample,producessensibleresults.
6. CONCLUDING REMARKS
in
The emphasis thispaper has been on developingthe Wtestin a formthatcan easilybe
programmedon a computer,fortheentirechosenrangeof samplesizes (3-2000). As faras I
know,thereis no otheromnibustestof normalitywhichcan be calculatedand assignedan
withoutrecourseto some table. D'Agostino
adequatelyexact level of statisticalsignificance
(1971),explainingwhyhe consideredit unwiseto tryto extendWto largesamples,said "Each
sample size requiresa new set of weightsforcalculatingW The proliferation of tables is
obvious and undesirable. . . and therewould stillbe the uninvitingproblemof findingthe
appropriatenull distributionof W." We have circumventedthese problemsby extending
Shapiro and Wilk's (1965) approximationsfor the weightsand by providinga good
normalizingtransformation forW up to samplesize 2000.
We aim to publish an algorithmembodyingthese methodsfor calculatingW and its
significancelevelin thisJournalin due course.

REFERENCES
BLOM,G. (1958).StatisticalEstimatesand Transformed Beta-variables.
New York: Wiley.
BOWMAN, K. 0. and SHENTON, L. R. (1975). Omnibustestcontoursfordeparturesfromnormalitybased on lb, and
b2. Biometrika, 62, 243-250.
D'AGosTINo,R. B. (1971).An omnibustestofnormality formoderateand largesamplesizes.Biometrika, 58, 341-348.
D'AGosTINo, R. B. and PEARSON, E. S. (1973). Tests for departurefromnormality.Empirical resultsfor the
distributions of b2 and lb,. Biometrika, 60, 613-622.
DAVIS,C. S. and STEPHENS, M. A. (1978). AlgorithmAS 128. Approximating thecovariancematrixof normalorder
statistics.Appl.Statist.,27, 206-212.
DE WET,T. and VENTER, J.H. (1972).Asymptotic ofcertaintestcriteriaofnormality.
distributions S. AfricaStatist.J.,
6, 135-150.
FILLIBEN, J.J.(1975). The probabilityplot correlationcoefficient testfornormality.Technometrics, 17, 111-117.
HARTER, H. L. (1961). Expectedvalues of normalorderstatistics.Biometrika, 48, 151-165.
JOHNSON, N. L. (1949). Systemsof frequency curvesgeneratedby methodsof translation.Biometrika, 36, 149-176.
PEARSON,E. S., D'AGOSTINO, R. B. and BOWMAN,K. 0. (1977). Tests for departure from normality: comparison of
powers.Biometrika, 64, 231-246.
RONA,R. J.and ALTMAN, D. G. (1977).NationalStudyofHealthand Growth:standardsofattainedheight,weightand
tricepsskinfoldin Englishchildren5 to 11 yearsold. Ann.Hum.Biol.,4, 501-523.
ROYSTON, J.P. (1982).Algorithm AS 177. Expectednormalorderstatistics (exactand approximate).Appl.Statist.,31,
161-165.
SARHAN, A. E. and GREENBURG, B. G. (1956). Estimationof location and scale parametersby orderstatisticsfrom
singlyand doublycensoredsamples.Part I. Ann.Math. Statist.,27, 427-451.
SHAPIRO,S. S. and FRANCIA, R. S. (1972).An approximateanalysisofvariancetestfornormality. J. Amer.Statist.Ass.,
67, 215-216.
124 APPLIED STATISTICS

SHAPIRO, S. S. and WILK, M. B. (1965). An analysisof variancetestfornormality 52,

(completesamples).Biometrika,
591-611.
(1968). Approximations forthe nulldistribution 10, 861-866.
of the Wstatistic.Technometrics,
SHAPIRO, S. S., WILK, M. B. and CHEN, H. J.(1968).A comparativestudyofvarioustestsofnormality.J. Amer.Statist.
Ass.,63, 1343-1372.
WICHMANN, B. A. and HILL, I. D. (1982). AlgorithmAS 183. An efficient and portablepseudo-randomnumber
generator.Appl.Statist.,31, 188-190.
WILK, M. B. and SHAPIRO, S. S. (1968). The joint assessmentof normalityof several independentsamples.
Technometrics, 10, 825-839.