ISO 2854-1976 (E)

TABLE A - Comparison of a mean with a given value (variance known)

Technical characteristics of the population studied:

Technical characteristics of the sample items:

Discarded observations:

Statistical data Calculations

Sample size:

n= x=
∑x=
n
Sum of the observed values:

∑x = [u1−α / n ]σ =
Given value:
[u1−α / 2 / n ]σ =
m 0=

Known value of the population variance:

σ 2=

Or standard deviation:

σ=

Significance level chosen:

α=

Results

Comparison of the population mean with the given value m0

Two-sided case:

| x − m0 |> [u1−α / n ] σ

The hypothesis of the equality of the population mean to the given value (null hypothesis) is rejected if:

One-sided cases:

a) The hypothesis that the population mean is not smaller than m0 (null hypothesis) is rejected if:

x < m0 − [u1−α / n ] σ

b) The hypothesis that the population mean is not greater than m0 (null hypothesis) is rejected if:

x > m0 + [u1−α / n ] σ
ISO 2854-1976 (E)

TABLE A´ - Comparison of a mean with a given value (variance unknown)

Technical characteristics of the population studied:

Technical characteristics of the sample items:

Discarded observations:

Statistical data Calculations

Sample size:

n= x=
∑x=
n
Sum of the observed values:
∑ ( x − x ) 2 = ∑ x 2 − (∑ x) 2 / n
∑x = n −1 n −1

Sum of the squares of the observed values: σ* =s=

∑ (x − x) 2
n −1
∑x2 =

Given value: [t1−α (ν ) / n ]s =

m 0=
[t1−α / 2 (ν ) / n ]s =
Degrees of freedom:

ν=n-1

Significance level chosen:

α=

Results

Comparison of the population mean with the given value m0

Two-sided case:

The hypothesis of the equality of the population mean to the given value (null hypothesis) is rejected if:

| x − m 0 |> [t1−α / 2 (ν ) / n ]s

One-sided cases:

a) The hypothesis that the population mean is not smaller than m0 (null hypothesis) is rejected if:

x < m 0 − [t1−α (ν ) / n ] s

b) The hypothesis that the population mean is not greater than m0 (null hypothesis) is rejected if:

x > m 0 + [t1−α (ν ) / n ] s
ISO 2854-1976 (E)

TABLE B – Estimation of a mean (variance known)

Technical characteristics of the population studied:

Technical characteristics of the sample items:

Discarded observations:

Statistical data Calculations

Sample size:

x=
∑x=
n=
n
Sum of the observed values:
[u1−α / n ]σ =
∑x =

Known value of the population variance: [u1−α / 2 / n ]σ =

σ 2=

Or standard deviation:

σ=

Confidence level chosen:

1-α =

Results

Estimation of the population mean m:

m*= x =

Two-sided confidence interval:

x − [u1−α / 2 / n ] σ < m < x + [u1−α / 2 / n ] σ

One-sided confidence intervals:

m < x + [u1−α / n ] σ

or

m > x − [u1−α / n ] σ
ISO 2854-1976 (E)

TABLE B´ - Estimation of a mean (variance unknown)

Technical characteristics of the population studied:

Technical characteristics of the sample items:

Discarded observations:

Statistical data Calculations

Sample size:

n= x=
∑x=
n
Sum of the observed values:
∑ (x − x) 2 = ∑ x 2 − (∑ x) 2 / n
∑x = n −1 n −1

Sum of the squares of the observed values: σ* =s=

∑ (x − x) 2
n −1
∑x2 =

Degrees of freedom: [t1−α (ν ) / n ]s =

ν=n-1
[t1−α / 2 (ν ) / n ]s =
Confidence level chosen:

1- α =

Results

Estimation of the population mean m:

m*= x =

Two-sided confidence interval:

x − [t1−α / 2 (ν ) / n ] s < m < x + [t1−α / 2 (ν ) / n ] s

One-sided confidence intervals:

m < x + [t1−α (ν ) / n ] s

or

m > x − [t1−α (ν ) / n ] s
ISO 2854-1976 (E)

TABLE C - Comparison of two means (variance known)

of population 1
Technical characteristics:
of population 2
Technical characteristics of in population 1
the sample items taken:
in population 2
Discarded observations: in sample 1
in sample 2
Statistical data Calculations
First Second
x1 =
∑ x1 =
sample sample n1
Size: n1= n2= ∑ x2
x2 = =
n2
Sum of the observed values: ∑x1= ∑x2=
σ 12 σ 22
Known value of the variances σ 12 = σ 22 = σd = + =
n1 n2
of the population: u1−α σ d =

Significance level chosen: u1−α / 2σ d =

α=

Results

Comparison of the two populations means:

Two-sided case:

The hypothesis of the equality of the means (null hypothesis) is rejected if:
| x1 − x 2 |> u1−α / 2σ d

One-sided cases:

a) The hypothesis that the first mean is not smaller than the second (null hypothesis) is rejected if:
x1 < x 2 − u1−α σ d

b) The hypothesis that the population mean is not greater than the second (null hypothesis) is rejected if:
x1 > x 2 + u1−α σ d
ISO 2854-1976 (E)

TABLE C´ - Comparison of two means (variance unknown, but may be assumed equal)

of population 1
Technical characteristics:
of population 2
Technical characteristics of in population 1
the sample items taken:
in population 2
Discarded observations: in sample 1
in sample 2
Statistical data Calculations
First Second
∑ x1
sample sample x1 = =
n1
Size: n1= n2=
∑ x2
x2 = =
Sum of the observed values: ∑x1= ∑x2= n2

∑ ( x1 − x1 ) + ∑ ( x 2 − x 2 ) =
Sum of the squares of the 2 2
∑ x12 = ∑ x 22 =
observe values:
∑ x1 + ∑ x 2 − (∑ x1 ) / n1 − (∑ x 2 ) / n 2 =
2 2 2 2

n1 + n 2 ∑ ( x1 − x1 ) 2 + ∑ ( x 2 − x 2 ) 2
Degrees of freedom ν= n1+ n2 -2= σ d* = s d = =
n1 n 2 n1 + n 2 − 2
Significance level chosen:
t1−α (ν ) s d =
α= t1−α / 2 (ν ) s d =

Results

Comparison of the two populations means:

Two-sided case:

The hypothesis of the equality of the means (null hypothesis) is rejected if:
| x1 − x 2 |> t1−α / 2 (ν ) s d

One-sided cases:

a) The hypothesis that the first mean is not smaller than the second (null hypothesis) is rejected if:
x1 < x 2 − t1−α (ν ) s d

b) The hypothesis that the population mean is not greater than the second (null hypothesis) is rejected if:
x1 > x 2 + t1−α (ν ) s d
ISO 2854-1976 (E)

TABLE D - Estimation of difference of two means (variances known)

of population 1
Technical characteristics:
of population 2
Technical characteristics of in population 1
the sample items taken:
in population 2
Discarded observations: in sample 1
in sample 2
Statistical data Calculations
First Second
x1 =
∑ x1 =
sample sample n1
Size: n1= n2= ∑ x2
x2 = =
n2
Sum of the observed values: ∑x1= ∑x2=
σ 12 σ 22
Known value of the variances σ 12 = σ 22 = σd = + =
n1 n2
of the population: u1−α σ d =

Confidence level chosen: u1−α / 2σ d =

1-α =

Results

Estimation of the difference of the two populations means m1 y m2:

(m1 − m 2 ) * = x1 − x 2 =

Two-sided confidence interval:

( x1 − x 2 ) − u1−α / 2σ d < m1 − m 2 < ( x1 − x 2 ) + u1−α / 2σ d

One-sided confidence interval:

m1 − m 2 < ( x1 − x 2 ) + u1−α σ d

or

m1 − m 2 > ( x1 − x 2 ) − u1−α σ d
ISO 2854-1976 (E)

TABLE D´ - Estimation of two means (variance unknown, but may be assumed equal)

of population 1
Technical characteristics:
of population 2
Technical characteristics of in population 1
the sample items taken:
in population 2
Discarded observations: in sample 1
in sample 2
Statistical data Calculations
First Second
∑ x1
sample sample x1 = =
n1
Size: n1= n2=
∑ x2
x2 = =
Sum of the observed values: ∑x1= ∑x2= n2

∑ ( x1 − x1 ) + ∑ ( x 2 − x 2 ) =
Sum of the squares of the 2 2
∑ x12 = ∑ x 22 =
observe values:
∑ x1 + ∑ x 2 − (∑ x1 ) / n1 − (∑ x 2 ) / n 2 =
2 2 2 2

n1 + n 2 ∑ ( x1 − x1 ) 2 + ∑ ( x 2 − x 2 ) 2
Degrees of freedom ν= n1+ n2 -2= σ d* = s d = =
n1 n 2 n1 + n 2 − 2
Confidence level chosen:
t1−α (ν ) s d =
1-α = t1−α / 2 (ν ) s d =

Results

Estimation of the difference of the two populations means m1 y m2:

(m1 − m 2 ) * = x1 − x 2 =

Two-sided confidence interval:

( x1 − x 2 ) − t1−α / 2 (ν ) s d < m1 − m 2 < ( x1 − x 2 ) + t1−α / 2 (ν ) s d

One-sided confidence interval:

m1 − m 2 < ( x1 − x 2 ) + t1−α (ν ) s d

or

m1 − m 2 > ( x1 − x 2 ) − t1−α (ν ) s d
ISO 2854-1976 (E)

TABLE E - Comparison of a variance or of a standard deviation with a given value

Technical characteristics of the population studied:

Technical characteristics of the sample elements:

Discarded observations:

Statistical data Calculations

Sample size:

n= (∑ x) 2
∑ (x − x) = ∑ x − =
2 2

Sum of the observed values: n

∑ (x − x)
2
∑x = =
σ 02
Sum of the squares of the observed values:
χ α2 (ν ) =
∑x2 =
χ 12−α (ν ) =
Given value:
χ α2 / 2 (ν ) =
σ 02 =
χ 12−α / 2 (ν ) =
Degrees of freedom:

ν=n-1
Significance level chosen:

α=

Results

Comparison of the population variance with the given value σ 02 :

Two-sided case:

The hypothesis that the population variance is equal to the given value (null hypothesis) is rejected if:

∑ (x − x) ∑ (x − x)
2 2
< χ α2 / 2 (ν ) or > χ 12−α / 2 (ν )
σ 02 σ 02

One-sided cases:

a) The hypothesis that the population variance is not larger than the given value (null hypothesis) is rejected if:

∑ (x − x)
2
> χ 12−α (ν )
σ 02

b) The hypothesis that the population variance is not smaller than the given value (null hypothesis) is rejected if:

∑ (x − x)
2
< χ α2 (ν )
σ 02
ISO 2854-1976 (E)

TABLE F - Estimation of a variance or of a standard deviation

Technical characteristics of the population studied:

Technical characteristics of the sample items:

Discarded observations:

Statistical data Calculations

Sample size:
(∑ x) 2
∑ (x − x) = ∑ x − =
2 2
n= n

∑ (x − x)
2
Sum of the observed values:
s2 = =
n −1
∑x =
∑ (x − x)
2
Sum of the squares of the observed values: =
χ α2 (ν )
∑x2 =
∑ (x − x)
2
=
Degrees of freedom: χ 12−α (ν )
ν=n-1 ∑ (x − x)
2
=
Confidence level chosen: χ α2 / 2 (ν )

1-α = ∑ (x − x)
2
=
χ 12−α / 2 (ν )

Results

Estimation of the population variance σ 2 :

(σ2) *=s2=

Two-sided confidence interval:

∑ (x − x) ∑ (x − x)
2 2
<σ 2 <
χ 12−α / 2 (ν ) χ α2 / 2 (ν )

One-sided confidence intervals:

∑ (x − x)
2
σ2 <
χ α2 (ν )

or

∑ (x − x)
2
σ2 >
χ 12−α (ν )
ISO 2854-1976 (E)

TABLE G – Comparison of two variances or of two standard deviations

of population 1
Technical characteristics:
of population 2
Technical characteristics of in population 1
the sample items taken:
in population 2
Discarded observations: in sample 1
in sample 2
Statistical data Calculations
First Second
sample sample ( ∑ x1 ) 2
∑ ( x1 − x1 ) = ∑ x1 − =
2 2
n1
Size: n1= n2=
(∑ x 2 ) 2
∑ (x 2 − x 2 ) = ∑ x 2 − =
2 2
Sum of the observed values: ∑x1= ∑x2=
n2
Sum of the squares of the ∑ x12 = ∑ x 22 = ∑ ( x1 − x1 )
2
observe values: s 12 = =
n1 − 1
Degrees of freedom: ν1= n1-1 ν2= n2-1
∑ (x 2 − x 2 )
2
s 22 = =
n2 − 1

Significance level chosen: F1−α (ν 1 , ν 2 ) = F1−α / 2 (ν 1 , ν 2 ) =

1 1
α= = =
F1−α (ν 2 , ν 1 ) F1−α / 2 (ν 2 , ν 1 )

Results

Comparison of the population variances:

Two-sided case:

The hypothesis of the equality of the variances (null hypothesis) is rejected if:

s12 1 s12
< or > F1−α / 2 (ν 1 ,ν 2 )
s 22 F1−α / 2 (ν 2 ,ν 1 ) s 22

One-sided cases:

a) The hypothesis that the first variance is not greater than the second (null hypothesis) is rejected if:

s12
> F1−α (ν 1 ,ν 2 )
s 22

b) The hypothesis that the first variance is not smaller than the second (null hypothesis) is rejected if:

s12 1
<
s 22 F1−α ν 2 ,ν 1 )
(
ISO 2854-1976 (E)

TABLE H – Estimation of the ratio of two variances or of two standard deviations

of population 1
Technical characteristics:
of population 2
Technical characteristics of in population 1
the sample items taken:
in population 2
Discarded observations: in sample 1
in sample 2
Statistical data Calculations
First Second
sample sample ( ∑ x1 ) 2
∑ (x1 − x1 ) = ∑ x1 − =
2 2
n1
Size: n1= n2=
(∑ x 2 ) 2
∑ (x 2 − x 2 ) = ∑ x 2 − =
2 2
Sum of the observed values: ∑x1= ∑x2= n2
∑ (x1 − x1 )
2
Sum of the squares of the ∑ x12 = ∑ x 22 = s 12 = =
observe values: n1 − 1
∑ (x 2 − x 2 )
2
Degrees of freedom: ν1= n1-1 ν2= n2-1 s 22 = =
n2 − 1
s 12 s 12
F1−α (ν 2 , ν 1 ) = F1−α / 2 (ν 2 , ν 1 ) =
Confidence level chosen: s 22 s 22
1 s 12 1 s 12
1-α = = =
F1−α (ν 1 , ν 2 ) s 22 F1−α / 2 (ν 1 , ν 2 ) s 22

Results

Estimation of the ratio of the two population variances σ 12 and σ 22 :

*
⎛ σ 12 ⎞ ⎛ s12 ⎞ ∑ ( x1 − x1 ) 2 /(n1 − 1)
⎜ ⎟ =⎜ ⎟=
⎜σ 2 ⎟ ⎜ s2 ⎟ ∑ ( x − x ) 2 /(n − 1)
⎝ 2 ⎠ ⎝ 2 ⎠ 2 2 2

Two-sided confidence interval:

1 s12 σ 12 s12
< < F −α (ν , ν )
F1−α / 2 (ν 1 ,ν 2 ) s 22 σ 22
1 / 2 2 1
s 22

One-sided confidence interval:

σ 12 s12 σ 12 1 s12
< F1−α (ν 2 ,ν 1 ) or >
σ 22 s 22 σ 22 F1−α (ν 1 ,ν 2 ) s 22