SECTION K: Chapter 65 of John Bird textbook

T MOTSIMA : 2023

PowePoint slides from Lombard C, van der Merwe L, Kele T & Mouton A also used
At the end of the chapter you will be able:
• to develop hypothesis testing methodology.
• to introduce the steps in the hypothesis testing
procedure.
• to describe hypothesis tests concerning population
mean, proportion and variance.

2
A hypothesis is a statement or claim that something is true
Two types of hypotheses (null and alternative)
– Null hypothesis is denoted by H0
• No effect or no difference
• Must be declared true/not true
– Alternative hypothesis is denoted by H1
• A hypothesis that differs from H0
Types of hypothesis tests
1. Two-tailed/ non-directional (H1 involves ≠)
2. 2. One-tailed/directional (H1 involves < or >)

At least : ≥ _________
At most : ≤ _________
Less/smaller/slower/lower/worse than: <
3
More/exceeds/greater/larger/higher/faster/better than : >
Types of hypothesis tests
1. Two-tailed/ non-directional (H1 involves ≠)
2. One-tailed/directional (H1 involves < or >)

H0 : µ = 5 H0 : µ ≤ 5 H0 : µ ≥ 5
H1 : µ ≠ 5 H1 : µ > 5 H1 : µ < 5

At least : ≥ minimum
At most : ≤ maximum
Less/smaller/slower/lower/worse than: <
More/exceeds/greater/larger/higher/faster/better than : >

4
Types of errors:
Decision Actual situation
H0 is true H0 is not true
Do no reject H0 Correct decision Type II error = β
Reject H0 Type I error = α Correct decision

α – probability of incorrectly rejecting H0 when it is true.

Type I error (α) –______________________________________________

Type II error (β) – _____________________________________________

_____________________________________________ 5
The six steps of hypothesis testing

Step 1: State the hypothesis

Step 2: State the value of α (α – level of significance/risk level
to tolerate)
Step 3: Calculate the test statistic
Step 4: Determine the critical value
Step 5: Make a decision
Step 6: Draw the conclusion

6
Hypothesis Testing Step 5
Make decision

Right tailed Two tailed Left tailed

H1: parameter > ? H1: parameter ≠ ? H1: parameter < ?

H0
H0
α H0
α/2 α/2 α

Critical regions are Reject H0 if the test statistic 7

highlighted in green falls within the critical region
Hypothesis test for Population Mean, n ≥ 30
- population need not be normally distributed
- sample will be approximately normal
Testing H0: μ = μ0 for n ≥ 30
Alternative Decision rule:
Test statistic
hypothesis Reject H0 if
H1: μ ≠ μ0 |z| ≥ Z1- α/2 x − 0
z=
H1: μ > μ0 z ≥ Z1- α  s 
 
H1: μ < μ0 z ≤ -Z1- α  n  8
Use σ if known
• Hypothesis test for Population Mean, n < 30
– If σ is unknown we use s to estimate σ
– We need to replace the normal distribution with the
t-distribution with (n - 1) degrees of freedom

Testing H0: μ = μ0 for n < 30

Alternative Decision rule:
Test statistic
hypothesis Reject H0 if
H1: μ ≠ μ0 |t| ≥ tn - 1;1- α/2 x − 0
t=
H1: μ > μ0 t ≥ tn-1;1- α  s 
 
 n
H1: μ < μ0 t ≤ -tn-1;1- α 9
• Hypothesis testing for Population proportion
number of successes x
– Sample proportion p = =
sample size n

– Proportion always between 0 and 1

Testing H0: π = p0 for n ≥ 30
Alternative Decision rule:
Test statistic
hypothesis Reject H0 if
H1: π ≠ p0 |z| ≥ Z1- α/2 p −
z=
H1: π > p0 z ≥ Z1- α  (1 −  )
n
H1: π < p0 z ≤ -Z1- α 10
. x − 0 x − 0
z = t=
 s   s 
   
 n 
 n

These assumptions are made when using Z-distribution for CI

i. the population standard deviation σ is known;
ii. the sample size is larger (n ≥ 30);
iii. the population of the random variable X is normally distributed

• If σ is unknown and n < 30, use the t-distribution;

• If σ is unknown and n ≥ 30, the t-distribution is close
enough to the z-distribution, use the z-distribution.
11
• Hypothesis testing for Population Variance
– Draw conclusions about variability in population
– Χ2 –distribution with (n - 1) degrees of freedom

Testing H0: σ2 = σ20

Alternative Test statistic Decision rule:
hypothesis Reject H0 if
Χ2 ≤ Χ2n-1;α/2 or
H1: σ2 ≠ σ20 (n − 1) s 2
 =
2 Χ2 ≥ Χ2n-1;1- α/2
H1: σ2 > σ20
 02 Χ2 ≥ Χ2n-1;1- α
H1: σ2 < σ20 Χ2 ≤ Χ2n-1;α 12
Example: State the null and alternative hypotheses

1. The nurse claims that the average weight of patients that came to the clinic
yesterday is less than 80kg.

2. The nurse claims that the average weight of patients that came to the clinic
yesterday is 80kg.

3. The nurse claims that the average weight of patients that came to the clinic
yesterday is different from 80kg.
Example: State the null and alternative hypotheses

4. The manager at the gym claims that the average weight of male clients who are
affiliated with his gym is less than 75.5kg.

5. The manager of a store claims that the proportion of customers who like to buy
fat cakes from their store is not 70%.

6. The sales manager claims that the proportion of clients who bought their product
is 30%.
Example
Sugar is packed in bags by an automatic machine. From a random sample of 36
that was selected throughout the day the mean mass of the contents of a bag is
1.003kg. The standard deviation on all bags packed is 0.01kg. Test at a 5% level of
significance whether the mean on all bags is 1kg.
n = 36;  = 1;  = 0.01
Step 1: H 0 :  = 1 vs H1 :   1 Step 2: α = 0.05
x − 0 1.003 − 1
Z = = = ________
Step 3: Test statistic:    0.01
 
 n  36

Step 4: Critical value: ___________

Step 5: Decision: Reject H0 if Z  Z  . Since Z = 1.8 = 1.8  1.96, do not reject H0.
1−
Step 6: At a 5% level of significance, there
2
is evidence to infer that the mean on all
sugar bags is 1kg. 15
Example
During research to determine the average calories contained in a hamburger, a
sample of 50 burgers were analysed and the average calorie count was 275 with a
standard deviation of 10. Assume calories contained in a hamburger are normally
distributed. Health experts claim that the average calorie count is in fact more than
280 calories in a burger. Test this claim using α = 0.05.
H 0 :   280
Step 1: Step 2: α = 0.05
H1 :   280

Step 3: Test statistic: Z = x − 0 = 275 − 280 = ____

 s  10
 
 n  50
Step 4: Critical value: ________

Step 5: Decision: Reject H0 if Z ≥ Z1- α. Since Z = -3.54 < 1.645, do not reject H0.

Step 6: Conclusion: At a 5% level of significance, there is evidence to infer that the

average calorie count is not more than 280 calories
16
 Hypothesis Testing
Example: Litres of cartoons of milk are advertised to
contain 1-litre of milk. To test this claim, the Consumer
Council of SA measured a random selection of 30 cartoons
from supermarket shelves. They found that the average fill
was 0.982 litres with a sample standard deviation of 0.068
litres. Assume that the fill of 1-litre milk cartoons is
approximately normally distributed. Test the claim that the
1-litre cartoons of milk are being under-filled. Use α=0.10.

17
Step 1: Step 2:  = ____

Step 3:

Step 4: − Z1− = _____ = ______

Step 5:

Step 6: At 5% level of significance, there is evidence to_____________

_________________________________________________
18
Hypothesis test for Population Mean, n ≥ 30
- population need not be normally distributed
- sample will be approximately normal
Testing H0: μ = μ0 for n ≥ 30
Alternative Decision rule:
Test statistic
hypothesis Reject H0 if
H1: μ ≠ μ0 |z| ≥ Z1- α/2 x − 0
z=
H1: μ > μ0 z ≥ Z1- α  s 
 
H1: μ < μ0 z ≤ -Z1- α  n  19
Use σ if known
• Hypothesis test for Population Mean, n < 30
– If σ is unknown we use s to estimate σ
– We need to replace the normal distribution with the
t-distribution with (n - 1) degrees of freedom

Testing H0: μ = μ0 for n < 30

Alternative Decision rule:
Test statistic
hypothesis Reject H0 if
H1: μ ≠ μ0 |t| ≥ tn - 1;1- α/2 x − 0
t=
H1: μ > μ0 t ≥ tn-1;1- α  s 
 
 n
H1: μ < μ0 t ≤ -tn-1;1- α 20
Example: When welder is in adjustment, its mean
cycle time to perform its task is 1.3250 minutes. An
incorrect mean operating time can disrupt the
efficiency of the other activities along the
production line. For a recent random sample of 21
jobs, the mean cycle time for the welder was
1.3229 minutes with the standard deviation of
0.0396 minutes. At a 1% level of significance, test
whether the machine needs some adjustments.
21
Step 1: Step 2: _____

Step 3: Test Statistic

Step 4: Critical value

Step 5: Decision

Step 6: Conclusion: ________________________________________

___________________________________________________
22
Example: The average weekly amount of defective items of steel at a
steel company is 21.7kg. A modification of the process is tried in an
attempt to eliminate the problem causing defective items. A sample
mean of 11.2kg and a standard deviation of 1.52kg resulted from a
15 week trial. At a 5% level of significance test whether the
modification process reduced the mean of defective items. The
population from which the sample was drawn is normally
distributed.

23
Step 1: Step 2: α = ____

Step 3: Test Statistic

Step 4: Critical value

Step 5: Decision

Step 6: Conclusion: ________________________________________

___________________________________________________
24
• Hypothesis testing for Population proportion
number of successes x
– Sample proportion p = =
sample size n

– Proportion always between 0 and 1

Testing H0: π = p0 for n ≥ 30
Alternative Decision rule:
Test statistic
hypothesis Reject H0 if
H1: π ≠ p0 |z| ≥ Z1- α/2 p −
z=
H1: π > p0 z ≥ Z1- α  (1 −  )
n
H1: π < p0 z ≤ -Z1- α 25
Example:
A union claims that 40% of labourers are in favour
of industrial action. During an opinion poll 106
labourers in a sample of 150 were in favour of
industrial action. Test the unions’ claim at a 10%
significance level.
x 106
n = 150; x = 106; p = = = 0.71
n 150

 = 0.10; 1 − = 1 − 0.05 = 0.95
2
Step 1: H0: π = 0.40
H1: π ≠ 0.40 26
Step 1: Step 2: α = ____

Step 3: Test Statistic

Step 4: Critical value

Step 5: Decision

Step 6: Conclusion: ________________________________________

___________________________________________________
27
• Hypothesis testing for Population Variance
– Draw conclusions about variability in population
– Χ2 –distribution with (n - 1) degrees of freedom

Testing H0: σ2 = σ20

Alternative Test statistic Decision rule:
hypothesis Reject H0 if
χ2 ≤ χ2n-1;α/2 or
H1: σ2 ≠ σ20 (n − 1) s 2
 =
2
χ2 ≥ χ2n-1;1- α/2
 02
H1: σ2 > σ20 χ2 ≥ χ2n-1;1- α
H1: σ2 < σ 2 0 χ2 ≤ χ2n-1;α 28
Example
The variation in the content of a 340ml can of beer should not be more than 10ml2.
To test the validity of this, 25 cans of beer revealed a variance of 12ml2. At a 10%
level of significance can we conclude that the variation in the content of the cans is
larger than 10ml2.

29
Step 1: Step 2: α = _______

Step 3: Test Statistic

Step 4: Critical value

Step 5: Decision

Step 6: Conclusion: ________________________________________

___________________________________________________
30
1. Lombard C, van der Merwe L, Kele T and Mouton S.
2012. Elementary Statistics for Business and
Economics.

2. John Bird. 2017. Higher Engineering Mathematics.

2. Trevor Wegner. 2006. Applied and Business Statistics:

Methods and Applications.

31