CHAPTER 9,10,11 MID-TERM (RSJ)

Conditions Table Formula

CHAPTER - 9
Hypothesis Case I. If the following three conditions are fulfilled: P-value x̄−μ σ
z=     where    σ x̄ =
Tests About µ : approach σ x̄ √n
1. The population standard deviation σ is known
σ Known 2. The sample size is small (i.e., n < 30) Z table
3. The population from which the sample is 1. State the null and
selected is alternative
normally distributed. hypothesis.
We use the normal distribution to perform a test of 2. Select the
hypothesis. distribution to use.
3. Calculate the p–
value.
Case II. If the following two conditions are fulfilled: 4. Make a decision.

1. The population standard deviation σ is known

2. The sample size is large (i.e., n ≥ 30) C-value
We use the normal distribution to perform a test of approach
hypothesis through the help of the C.L Theorem.
Z table
1. State the null and
Case III. If the following three conditions are fulfilled: alternative
hypotheses.
1. The population standard deviation σ is known 2. Select the
2. The sample size is small (i.e., n < 30) distribution to use.
3. The population from which the sample is 3. Determine the
selected is rejection and
not normally distributed (or its distribution is nonrejection regions.
unknown). 4. Calculate the value
We use a nonparametric method to perform a test of of the test statistic.
hypothesis. 5. Make a decision.
Hypothesis Case I. If the following three conditions are fulfilled: P-value x̄−μ s
t=    where    s x̄ =
Tests About µ :
1. The population standard deviation σ is not known
approach s x̄ √n
σ Unknown 2. The sample size is small (i.e., n < 30) t table
3. The population from which the sample is selected df = n – 1
is
normally distributed.
C-value
Case II. If the following two conditions are fulfilled: approach
t table
1. The population standard deviation σ is not known
2. The sample size is large (i.e., n ≥ 30)

In both these cases we use a t-distribution to conduct

the test.

Case III. If the following three conditions are fulfilled:

1. The population standard deviation σ is not known

2. The sample size is small (i.e., n < 30)
3. The population from which the sample is selected
is not
normally distributed (or the shape of its
distribution is
unknown).
 In this case we use a non-parametric method to
conduct the test.

CHAPTER - 10
Conditions Table Formula

√
Independent If these 3 conditions are satisfied: We can then σ 21 σ 22
Samples: 1.The two samples are independent conclude that 1. σ x̄ − x̄ = +
σ 1 and σ 2 2.The standard deviations σ1 and σ2 of The sampling
1
n1 n2
2

known the two populations are known/given distribution of ( x̄ − x̄ )−(μ1−μ2 )

2. z= 1 2
3. At least one of the following two is ( x 1- x 2 ¿ σ x̄ −x̄ 1 2

conditions is fulfilled: (approximately)

i. Both samples are large (i..e., n1 ≥ normally
30 and n2 ≥ 30) distributed and
ii. If either one or both sample sizes we use the
are small, then both populations from normal
which the samples are drawn are distribution/z
normally distributed. table

√
Independent If these 3 conditions are satisfied: We can then (n1−1)s1 +(n2−1)s2
2 2

Samples: 1. The two samples are independent conclude that 1. s p=

σ 1 and σ 2 2. The standard deviations σ1 and σ2 of We will use the n 1+ n2−2
unknown and
assumed to be
equal
the two populations are unknown but
are assumed to be equal.
3. At least one of the following two
t-distribution
in order to
conduct the
2.
s x̄ −x̄ =s p
1 2

3. df = n1 + n2 – 2
√ 1 1
+
n1 n 2

conditions is fulfilled: hypotheses

i. Both samples are large (i..e., n1 ≥ testing.
( x̄ 1− x̄ 2)−( μ1−μ 2)
30 and n2 ≥ 30) t=
4. s x̄ −x̄
ii. If either one or both sample sizes 1 2

are small, then both populations from

which the samples are drawn are
normally distributed.

( )
Independent If these 3 conditions are satisfied: We can then s 21 s 22
2

Samples: 1. The two samples are independent conclude that +

n1 n 2
σ 1 and σ 2 2. The standard deviations σ 1 and σ 2 of The t df =

( ) ( )
1. 2 2 2 2
unknown and the two populations are unknown and distribution is s1 s2
assumed to be unequal used to with a n1 n2
unequal 3. At least one of the following two special degrees +
conditions is fulfilled: of freedom n1−1 n2−1
Both samples are large (i.e., n1 formula which
√
2 2
I. s s
≥ 30 and n2 ≥ 30) is provided in 2. s x̄ − x̄ = 1 + 2
1
n 1 n2
2

II. If either one or both sample the next slide

sizes are small, then both
populations from which the ( x̄ 1− x̄ 2)−(μ1−μ 2)
3. t =
samples are drawn are s x̄ − x̄ 1 2

normally distributed.
Hypothesis Fulfilling the following conditions If so then we 1. H0: μd = 0
Testing About 1. The standard deviation dofσ can use the t-
μd the population of paired distribution to
differences is unknown test a
 2. At least one of the following hypothesis
2. d̄=
∑d
two conditions is fulfilled: about μd n
 The sample size is large
(i.e., n ≥ 30 ) σd
 If the sample size is 3. σ d̄=
small, then the √n
population of paired
differences is normally
sd
4. sd̄ =
distributed √n
5. sd = √∑ d −¿ ¿ ¿ ¿ ¿
2

d̄−μd
6. t =
sd̄

7. df = n – 1
Two  As was indicated in Chapter 9, Z table n1 ^
p1 + n2 ^p2
Population in the case of proportion, the 1. p=
n1 + n2
Proportions for sample is large if np and nq are 2. q=1− p

√
Large and both greater than 5.
Independent
Samples
 In the case of two samples,
both sample sizes are large if
3. s ^p − ^p = pq
1 2
1 1
+
n1 n2( )
n1 p1 , n1 q1 , n2 p2 ,n2 q 2are all (^ p1− ^p2 ) −( p1− p2 )
greater than 5. 4. z=
s
 However- Since we are not x x
aware of the population 5. ^p1= 1 ^p2= 2
n1 n2
proportions in this chapter -we
use the sample proportion
values. Hence forth we check
for n1 ^p1, n1 q^ 1 , n2 ^p2,n2 q^ 2are
all greater than 5.
 Whereq^ 1=(1− ^p 1) and
q^ 2=(1− ^p2)

Conditions Table Formula

CHAPTER 11
An experiment with the following chi-square 1. Area in the right tail = 1-
characteristics is called a multinomial distribution Area in the left tail
experiment. 2. df = k – 1
1. The experiment consists of n 3. E = np
identical trials (repetitions). 4.
χ =∑ ¿ ¿ ¿
2
2. Each trial results in one of k
possible outcomes (or
categories), where k > 2. O = observed frequency for a category
3. The trials are independent.  E = expected frequency
for a category = np
 4. The probabilities of the various
outcomes remain constant for each
trial.

Contingency The number of cells in a contingency chi-square 1.

Tables : Test of table is obtained by multiplying the distribution (Row total)(Column total)
Independence number of rows by the number of E=
Sample size
& Test of columns. 2. df = (R – 1)(C – 1)
Homogeneity The subjects that belong to a cell of a 3. χ =∑ ¿ ¿ ¿
2

contingency table possess two

characteristics.

In a test of homogeneity, we test if two

(or more) populations are
homogeneous (similar) with regard to
the distribution of a certain
characteristic.

Hypothesis
Test about the
Population
Variance

Prepared by

MD AL ALIF HOSSAIN

Denotations (chapter 10)

𝒅 = paired difference, the difference between the two data values for each element of the two samples

𝒅 ̅ = the mean of the paired differences for the sample (we usually calculate from the given data)

𝒏 = the number of paired differences

𝝁_𝒅 = the mean of the paired differences for the population

𝝈_𝒅 = the standard deviation of the paired differences for the population , which is usually never known
𝒔_𝒅 =the standard deviation of the paired differences for the sample (we usually calculate from the
given data)

(𝒏-1) = degrees of freedom