Applied Business Statistics, 7th ed.

Chapter 9

Statistical Inference:
Hypothesis Testing
for Single
Populations
 Chapter Goals
◼ After completing this chapter, you should be able to:

◼ Formulate null and alternative hypotheses for applications

involving a single population mean variance is known and
unknown) or proportion

◼ Formulate a decision rule for testing a hypothesis

◼ Know how to use the test statistic, critical value, and p-value
approaches to test the null hypothesis
 Basic steps when testing hypotheses

◼ 1. Establish the hypotheses

◼ The null hypothesis denoted H0 is a claim about the

parameter. You want to prove the opposite

◼ The alternative hypotheses denoted Ha is the researcher

hypothesis, is the complement for the null.

◼ Could take one of the following forms :

 The null and alternative hypotheses
◼ For a single population mean , it could take one of the following forms: where is a pre-specified
value 0
1. H 0 :  =  0 or 2. H 0 :    0 or 3. H 0 :   0
H A :   0 H A :   0 HA :  

◼ For a single population proportion , where a pre specified proportionP0

◼
1. H 0 : P = P0 or 2. H 0 : P  P0 or 3. H 0 : P  P0
H A : P  P0 H A : P  P0 H A : P  P0

◼ Where (1) : two tail test non-directional

◼ (2) one tail to the left

◼ (3) one tail to the right

Step 2:
 Step 3

◼ 3. Select the test statistics:

The general form for the test statistic is:

Po int estimate − value under the null hypothesis
test statistic =
s tan dard error for the estimator
 Step 4: Formulate a decision Rule
(a) When the alternative hypothesis has ≠ sign “ not equal”

the test non-directional. H 0 :  = 0 H a :   0

α is divided on both directions, α/2 in each tail of

the sampling distribution ,
(1- α)
called critical values Critical Critical
Value
Value
 Step 4: Formulate a decision Rule

◼ (b) When Ha has > sign , the test directional, and the
area of rejection is to the right of the sampling distribution.
H 0 :  = 0 H a :   0
◼ The test is called one tail test.
◼ In this situation the α is put in the right tail of the sampling
distribution.
(1- α)

Critical Value
Step 4: Formulate a decision Rule

H 0 :  = 0 H a :   0

(1- α)

Critical Value
 When sampling distribution is normal

◼ Use the following values for critical values

 When sampling distribution is not normal

◼ Use t distribution with (n-1) degrees of freedom

◼ for α= .05 and non-directional test look it up using column .05/2=.025 .

◼ For α= .05 and directional test look it up using column .05 ,
◼ if the test is directional to the right then obtained tabled t-value is
positive
◼ if the test is directional to the left of the obtained tabled t-value takes a
negative sign.
Step 5: Make a decision
 Critical value Approach
Level of significance =  Represents
critical value

Rejection
Lower tail test 0 region is
shaded


Upper tail test 0

/2 /2

Two tailed test 0

 Cases to be studied
Po int estimate − value under the null hypothesis
test statistic =
s tan dard error for the estimator

X
z= −  and  = 
x x n
 Po int estimate − value under the null hypothesis
Examples case Problem 3.3
Test statistic =
s tan dard error for the estimator

x − 0 28.5 − 30
Z= = = −1.37
x 8.5
60

◼
Problem 3.2

x = 1,515 n = 120  = 100  = .10

x − 0 1515 − 1500
Z= = = 1.65
x 100
120
 x−
1515 − 1500
z= = = 1.65
 100
n 120

Rejection Region
Non Rejection Region

 Critical Value 1.65

Test statistic=1.65

Decision: since 1.65 is in the rejection area then Reject the null
hypothesis ,thus µ > 1,500
 
xz
n
100
1515  1.96
120
1515  17.89
P(1497 .11    1532 .89) = .95
 Problem 3

Rejection Region Non Rejection Region

=0
-2.33
 Ho: µ = 82 Ha: µ < 82
= 78.125 n = 32 s = 9.184  = .01

◼ The test statistic

x − : 78.125 − 82
Z= = = −2.39
s 9.184
n 32

Rejection Region Non Rejection Region

Decision: -2.39 -2.33 =0

reject the null
hypothesis
 Problem 5

◼ Ho: µ = 82 Ha: µ < 82

x = 78.125 n = 32 s = 9.184  = .01
x −  0 78.125 − 82
Z= = = −2.39
sx 9.184
◼ Critical value: -2.33 32

◼ Decision : Reject the null since -2.39 is less than -2.33.

◼ Conclusion: the population mean is less than 82.

 Po int estimate − value under the null hypothesis
test statistic =
Case 3
s tan dard error for the estimator

t= X − 
s/ n
 Rejection Region
Non Rejection Region
-
-2.718 1.46 =0

◼
x = 42.167 n = 12 s = 9.124  = .01

42.167 − 46 78.125 − 82
t= = = −1.46
sx 9.124
12
◼

X  t11,.025  s x
9.124
= 42.16  2.201  = 42.16  2.634
12
 Hypothesis Test of p

P = P0 P = P0 P = P0
P  P0 P  P0 P  P0
 two conditions must be met

When

The sample was randomly selected from the population

a) np0>= 5 and nq0 >=5.

Then : Samples are considered large and The critical values are
values

from the normal distribution

 Po int estimate − value under the null hypothesis
test statistic =
s tan dard error for the estimator

◼
P0  (1 − P0 )
 P =
n


P − P0
Z=
 P
 Problem 7: proportion

◼  x 58
P= = = .58
n 100
P0  (1 − P0 ) .65(1 − .65)
 P = = = .0476
n 100


P − P0 .58 − .65
Z= = = −1.47
 P .0476

Rejection Region Non Rejection Region

=0
-2.33 -1.47
◼

H 0 : P  .60 H a : P  .60

Decision: Reject the null

 1645 .60  .40 hypothesis
P= = .8225  P = = .01095
2000 2000
.8225 − .60
z= = 20.375
.01095 Conclusion : the percentage of
students who favor e-learning is more
than 60% .
Chapter testing hypotheses formulas