SLOVENSKI STANDARD

SIST ISO 2602:1996

01-september-1996

Statistical interpretation of test results - Estimation of the mean - Confidence

interval

Statistical interpretation of test results -- Estimation of the mean -- Confidence interval

iTeh STANDARD PREVIEW

Interprétation statistique de résultats d'essais -- Estimation de la moyenne -- Intervalle de
confiance (standards.iteh.ai)
SIST ISO 2602:1996
Ta slovenski standard je istoveten z: ISO 2602:1980
https://standards.iteh.ai/catalog/standards/sist/2e27beef-58d3-4cef-bd9a-
cf995e0d3f7f/sist-iso-2602-1996

ICS:
03.120.30 8SRUDEDVWDWLVWLþQLKPHWRG Application of statistical
methods

SIST ISO 2602:1996 en

2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
 SIST ISO 2602:1996

iTeh STANDARD PREVIEW

(standards.iteh.ai)
SIST ISO 2602:1996
https://standards.iteh.ai/catalog/standards/sist/2e27beef-58d3-4cef-bd9a-
cf995e0d3f7f/sist-iso-2602-1996
 SIST ISO 2602:1996

International Standard 2602

INTERNATIONAL ORGANIZATION FOR STANDARDIZATION.MElKt(AYHAPOAHAR OPTAHM3ALU4R n0 CTAH,QAPTl43A~bIWORGANISATION INTERNATIONALE DE NORMALISATION

Statistical interpretation of test results - Estimation of

the mean - Confidence interval
Interpretation statistique de rksultats d’essais - Estimation de Ia moyenne - Intervalle de confiance

Second edition - 1980-02-15

iTeh STANDARD PREVIEW

(standards.iteh.ai)
SIST ISO 2602:1996
https://standards.iteh.ai/catalog/standards/sist/2e27beef-58d3-4cef-bd9a-
cf995e0d3f7f/sist-iso-2602-1996

UDC 519.25 : 620.113 Ref. No. ISO 2602-1980 (E)

Descriptors : statistical analysis, statistical tests, estimation, test results, mean, variance (statistics).

Price based on 5 pages

 SIST ISO 2602:1996

Foreword
ISO (the international Organization for Standardization) is a worldwide federation of
national Standards institutes (ISO member bodies). The work of developing Inter-
national Standards is carried out through ISO technical committees. Every member
body interested in a subject for which a technical committee has been set up has the
right to be represented on that committee. International organizations, governmental
and non-governmental, in liaison with ISO, also take part in the work.

Draft International Standards adopted by the technical committees are circulated to

the member bodies for approval before their acceptance as International Standards by
the ISO Council.

International Standard ISO 2602 was developed

Applica tions of s ta tis ticai me thods.
iTeh STANDARD PREVIEW
by Technical Committee ISO/TC 69,

(standards.iteh.ai)
This second edition was submitted directly to the ISO Council, in accordance with
clause 5.10.1 of part 1 of the Directives for the technical SIST work ISO
of ISO. lt cancels and
2602:1996
replaces the first edition (i.e. ISO 2602-1973), which has been approved by the member
bodies of the following
https://standards.iteh.ai/catalog/standards/sist/2e27beef-58d3-4cef-bd9a-
countries :
cf995e0d3f7f/sist-iso-2602-1996
Australia India Portugal
Austria Ireland Romania
Belgium Israel South Africa, Rep. of
Czechoslovakia Italy Sweden
Egypt, Arab Rep. of Japan Switzerland
France Netherlands Thailand
Germany, F. R. New Zealand United Kingdom
Hungary Poland USSR

No member body had expressed disapproval of the document.

0 International Organkation for Standardkation, 1980

Printed in Switzerland
 SIST ISO 2602:1996

INTERNATIONAL STANDARD ISO 2602-1980 (EI

Statistical interpretation of test results - Estimation of

the mean - Confidence interval
Second edition

0 Introduction of the population of results that would be obtained from a very

large number of determinations, carried out under the same
The scope of this International Standard is limited to a special conditions. In the case of items with a variability, this Interna-
question. lt concerns only the estimation of the mean of a nor- tional Standard assumes that the individuals on which the
mal population on the basis of a series of tests applied to a ran- determinations are carried out constitute a random Sample
dom Sample of individuals drawn from this population, and from the original population and may be considered as indepen-
deals only with the case where the variance of the population is dent.
unknown. lt is not concerned with the calculation of an interval
containing, with a fixed probability, at least a given fraction of The interval so calculated is called the confidence interval for
the population (statistical tolerante limits). the mean. Associated with it is a confidence Ievel (sometimes
termed a confidence coefficient), which is the probability,
lt is recalled that ISO 2854 relates to the following collection of usually expressed as a percentage, that the interval does con-
Problems (including the Problem treated in this International tain the mean of the population. Only the 95 % and 99 % levels
Standard) : are provided for in this International Standard.

-
iTeh STANDARD PREVIEW
estimation of a mean and of the differente between two
(standards.iteh.ai)
means (the variances being either known or unknown);
1 Scope
- comparison of a mean with a given value and of two
SIST ISO 2602:1996
means with one another (the variances being either known This International Standard specifies the statistical treatment of
or unknown, but equal);
https://standards.iteh.ai/catalog/standards/sist/2e27beef-58d3-4cef-bd9a-
test results needed to calculate a confidence interval for the
cf995e0d3f7f/sist-iso-2602-1996
mean of a population.
- estimation of a variance and of the ratio of two
variances;

- comparison of a variance with a given value and of two 2 Field of application

variances with one another.
The test results are expressed by measurements of a con-
The test methods generally provide for several determinations tinuous Character. This International Standard does not cover
which are carried out : tests of a qualitative Character (for example presence or
absence of a property, number of defectives, etc.).
- on the same item (where the test is not destructive);
The probability distribution taken as a mathematical model for
- on distinct portions of a very homogeneous product (a the total population is a normal distribution for which
liquid, for example); Parameters, mean m and Standard deviation a, are unknown.

- on distinct items sampled from an aggregate with a cer- The normality assumption is very widely satisfied : the distribu-
tain amount of variability. tion of the results obtained under test conditions is generally a
normal or nearly normal distribution.
In the first two cases, the deviations between the results ob-
tained depend only upon the repeatability of the method. In the lt may, however, be useful to check the validity of the assump-
third case, they depend also on the variability of the product tion of normality by means of appropriate methods’).
itself.
The calculations may be simplified by a Change of the origin or
The statistical treatment of the results allows the calculation of the unit of the test results but it is dangerous to round off these
an interval which contains, with a given probability, the mean results.

1) This subject is in preparation.

 SIST ISO 2602:1996

ISO 2602-1980 (E)

lt is not permissible to discard any observations or to apply any The midpoint of class i is designa ted by yi. The mean m is then
corrections to apparently doubtful observations without a estimated by the weighted mean of all midpoin ts of classes :
justification based on experimental, technical or other evident
grounds which should be clearly stated.
_ 1 k
Y = y C niYi
The test method may be subject to systematic errors, the deter-
i= 1
mination of which is not taken into consideration here. lt
should be noted, however, that the existente of such errors
may invalidate the methods which follow. In particular, if there
is an unsuspected bias the increase of the Sample size n has no
influence on bias. The methods that are treated in ISO 2854
may be useful in certain cases for identifying systematic errors.
6 Confidence interval for the mean

The confidence interval for the population mean is caiculated

from the estimates of the mean and of the Standard devia tion .
3 References
The alternative method of calculating the confidence interval by
ISO 2854, Statistical treatment of data - Problems of estima- use of the range is given in the annex.
tion and tests of means and variances.

ISO 3534, Statistics - Vocabulary and symbols.

61. Estimation of the Standard deviation

4 Definitions and Symbols

6.1.1 Case of ungrouped results
The vocabulary and Symbols used in this International iTeh STANDARD PREVIEW
Standard
The estimate of the Standard deviation o, calculated from the
are in conformity with ISO 3534.
(standards.iteh.ai) squares of the deviations from the arithmetic mean, is given by
the formula :

SIST ISO 2602:1996

J
5 Estimation of the mean https://standards.iteh.ai/catalog/standards/sist/2e27beef-58d3-4cef-bd9a-
1 *
S= (Xi - x7*
cf995e0d3f7f/sist-iso-2602-1996
n-l z
5.1 Case of ungrouped results i= 1

After the discarding of any doubtful results, the series com-

prises n measurements Xi (where i = 1, 2, 3, . . . . n), some of
which may have the same value. Xi is the value of the ith measurement (i = 1, 2, 3, . . . . n);

The mean m of the underlyi ng normal distribu tion is estimated y1 is the total number of measurements;
by the arithmetic mean x of the n results : -
x is the arithmetic mean of the n measurements, calculated
as in clause 5.1.
- 1 *
x = - xi
n t: For ease of calculation, the use of the following formula is
i= 1
recommended :

5.2 Case of results grouped in classes

When the number of results is sufficiently high (above 50 for

example), it may be advantageous to group them into classes
of the same width. In certain cases, the results may also have
been directly obtained grouped into classes.

The frequency of the ith class, i.e. the number of results 6.1.2 Case of grouped results
class i, is deno ted by ni.
In the case of grouping by classes, the formula for the estimate
The number of classes being denoted by k, we have : of the Standard deviation is written :

1 k
n = ‘i S= ni (yi - fl*
c n-l c
i=l i= 1
 SIST ISO 2602:1996

ISO 2602-1980 (E)

For ease of calculation, the use of the following formula is or

recommended :
*o 95
K?>Z--LS
Al-n

b) at the confidence level 99 % :

where *o 99
m<x+is
d- n
Yj is the mid-Point of the ith class (i = 1, 2, 3, . . . . k);

k is the number of classes;

- .
y IS the weighted mean of all mid-Points of classes - .
calculated as in sub-clause 5.2. with X, rf necessary, replaced by y, In the case of results
grouped in classes.
In the case of grouped results, the calculated value of s should
be corrected (“Sheppard’s correction”). As this correction is of The values t 0,975’ *0,995’ *0,95’ *0,99 are those of Student’s t
secondary importante, it has not been mentioned here. distribution with v = n + 1 degrees of freedom.

These values are given in table 1.

6.2 Confidence interval for the mean
This table gives also the values of ratios
Fora Chosen confidence level (95 % or 99 %), according to the
specific case, a two-sided
has to be determined.
iTeh STANDARD PREVIEW
or a one-sided confidence interval
*0,975 ~*0,995 -*0,95 -*0,99
(standards.iteh.ai) Ji’@‘fi’Jz-
6.2.1 Two-sided confidence interval
When values of n are greater than 60, it is preferable to
SIST ISO 2602:1996 120
The two-sided confidence interval for the population mean is calculate the value of t by linear interpolation from -J- using
defined by the following
https://standards.iteh.ai/catalog/standards/sist/2e27beef-58d3-4cef-bd9a-
double inequality :
cf995e0d3f7f/sist-iso-2602-1996
table 2.
a) at the confidence level 95 % :
Example :

- *0,975 - *o 975 n = 250

x-- S<rn<x+-‘-,
ZTn Al-n
120
= o,a
b) at the confidence level 99 % n

- *0,995 - *0,995 t0.995 = 2,576 + 0,48 + (2,617 - 2,576)

x- -s<m<x+ -S
n = 2,596
d- d-n

6.2.2 One-sided confidence interval

. 7 Presentation of the results
The one-sided confidence interval for the population mean is
defined by one or other of the following inequalities : 7.1 Give the expression of the mean according to 5.1 or 5.2.

a) at the confidence level 95 % :

7.2 Express the confidence interval in the form of the double
inequality of 6.2.1 or one of the inequalities of 6.2.2, stating the
- *o 95
rn<x+Ls confidence level (95 % or 99 %). Indicate the number of results
d- n discarded as being doubtful and the reasons for discarding.
 SIST ISO 2602:1996

ISO 2602-1980 (E)

Table 1 - Values of t, _ cyand of the ratio t, _ ,/fi

Confidence level Confidence level

Two-sided case Bne-sided case

95 % 99 % 95 % 99 %
-~- ~~
*0,975
--- *0,995 ~-*0,95 -~*0,99
n
d- n d--n WK d--n
8,985 45,013 4,465 22,501
2,484 5,730 1,686 4,021
1,591 2,920 1,177 2,270
1,242 2,059 0,953 1,676
1,049 1,646 0,823 1,374
7 0,925 1,401 0,734 1,188
8 0,836 1,237 0,670 1,060
9 0,769 1,118 0,620 0,966
IO 0,715 1,028 0,580 0,892
11 0,672 0,956 0,546 0,833
12 0,635 0,897 0,518 0,785
13 0,604 0,847 0,494 0,744
14 0,577 0,805 0,473 0,708
15 0,554 0,769 0,455 0,668
16 0,533 0,737 0,438 0,651
17 0,514 0,708 0,423 0,627
18 0,497 0,683 0,410 0,605
19 0,482 0,660 0,398 0,586
20 0,468 0,640 0,387 0,568
iTeh STANDARD PREVIEW 21 0,455 0,621 0,376 0,552
22 0,443 0,604 0,367 0,537
(standards.iteh.ai) 23
24
0,432
0,422
0,588
0,573
0,358
0,350
0,523
0,510
25 0,413 0,559 0,342 0,498
SIST ISO 2602:1996 26 0,404 0,547 0,335 0,487
https://standards.iteh.ai/catalog/standards/sist/2e27beef-58d3-4cef-bd9a-
27 0,396 0,535 0,328 0,477
cf995e0d3f7f/sist-iso-2602-1996
28 0,388 0,524 0,322 0,467
29 0,380 0,513 0,316 0,658
30 0,373 0,503 0,310 0,449
40 0,320 0,428 0,266 0,384
50 0,284 0,379 0,237 0,340
60 0,258 0,344 0,216 0,309

Table 2

120
n *0,975 *0,995 *0,95 *0,99
n
\
60 2 2,000 2,664 1,673 2,393
120 1 1,980 2,617 1,658 2,358
Co 0 1,960 2,576 1,645 2,326