You are on page 1of 39

One-Sample

Tests of Hypothesis
1. Statistical Techniques in Business &
Economics by Lind, Marchal, and
Mason
2. Statistics for Managers by Levine,
Stephan, Krehbiel, Berenson

1 lilik for mm 01/08/22


Introduction
As the operations manager of the plant,
you are in charge of monitoring the
amount of cereal contained in the boxes.
You are concerned with evaluating
whether or not the cereal filling process
is working to ensure that on average,
368 grams of cereal are in each box. A
random sample of 25 boxes is selected
and the boxes are weighted to determine
how close the fill weights are to the
company’s specification of a mean of
368 grams.
2 lilik for mm 01/08/22
Cont’d…
You hope that the process is working
properly, but the sampled boxes may contain
too little cereal or perhaps too much. By
analyzing the differences between the weights
obtained from the sample and the 368-gram
expectation obtained from the company’s
specification, you need to decide whether the
average fill in the entire process is equal to
368 grams or is not equal to 368 grams.
3 lilik for mm 01/08/22
Hypothesis Testing Methodology
Hypothesis testing typically begins
with some theory, claim, or assertion
about a particular parameter of a
population
About the cereal example is the
process working properly, meaning
that the mean fill is 368 grams

4 lilik for mm 01/08/22


Definitions
Hypothesis: A statement about the value of a
population parameter developed for the purpose
of testing.
Hypothesis testing: A procedure, based on
sample evidence and probability theory, used to
determine whether the hypothesis is a
reasonable statement and should not be rejected,
or is unreasonable and should be rejected.

5 lilik for mm 01/08/22


Step Procedure for Hypothesis Testing

S t e p 1 : S t a t e n u ll a n d a lt e r n a t e h y p o t h e s e s

S t e p 2 : S e le c t a le v e l o f s ig n ific a n c e

S t e p 3 : F o r m u la t e a d e c is io n r u le

S t e p 4 : I d e n t ify a n d d e t e r m in e t h e t e s t s t a t is t ic b a s e d o n s a m p le d a t a

S t e p 5 : M a k e a d e c is io n / c o n c lu s io n

D o n o t r e je c t n u ll R e je c t n u ll a n d a c c e p t a lt e r n a t e

01/08/22 lilik for mm 6


State Null and Alternate Hypotheses
Null Hypothesis H0:
The hypothesis that the population
parameter is equal to the company
specification
A statement about the value of a population
parameter.
Alternative Hypothesis H1 or Ha: A statement
that is accepted if the sample data provide
evidence that the null hypothesis is false.
7 lilik for mm 01/08/22
State Null and Alternate Hypotheses

Ho H1 or Ha Test of Significance
1 =  Two-tailed

2  > One-tailed (right side)


=
3  < One-tailed (left side)
=

01/08/22 lilik for mm 8


Select a level of significance
Level of Significance: The probability of
rejecting the null hypothesis when it is
actually true.
Type I Error () : Rejecting the null
hypothesis when it is actually true.
Type II Error () : Accepting the null
hypothesis when it is actually false

9 lilik for mm 01/08/22


Select a level of significance
Researcher
Null Accepts Ho Reject Ho
Hypothesi
s

Ho is true Correct Type I Error


Decision ()
Ho is Type II Error Correct
false () Decision
01/08/22 lilik for mm 10
Formulate the Decision Rule
Critical value: The dividing point between the
region where the null hypothesis is rejected and
the region where it is not rejected.
p-Value: the probability, assuming that the null
hypothesis is true, of getting a value of the test
statistic at least as extreme as the computed
value for the test.
If the p-value is smaller than the significance level,
H0 is rejected.
If the p-value is larger than the significance level,
H0 is not rejected.
11 lilik for mm 01/08/22
Select The Test Statistic
Test statistic: A value, determined
from sample information, used to
determine whether or not to reject the
null hypothesis.

12 lilik for mm 01/08/22


Testing for the Population Mean
If the population standard deviation is
known, the test statistic is the standard normal
distribution and its value determined from :
X 
z
 / n
The numerator measures the distance between the
observed sample mean and the hypothesized mean
µ. The denominator is the standard error of the
mean, so Z represents the difference between and
µ in standardized units X

01/08/22 lilik for mm 13


Testing for the Population Mean
If the population standard deviation is not known,
but there are at least 30 observations in the sample
the test statistic is the standard normal distribution
and its value determined X from
 :
z
s/ n
 If the population standard deviation is not known, but
there are fewer than 30 observations in the sample the test
statistic is the t distribution and its value determined  :
Xfrom
t
s/ n

01/08/22 lilik for mm 14


Example
1. State H0 and H1
H0: µ = 368
H1: µ ≠ 368
2. Select level of significance
LOS = 5%

15 lilik for mm 01/08/22


3. Formulate a decision rule

Reject H0 if t> +2.06


Or t< - 2.06
Otherwise do not reject H0

Region of rejection
Region of rejection

0.025 0.025
µ

Crit value = -2.06 Crit value = +2.06

16 lilik for mm Region of non-rejection 01/08/22


4. Identify and determine the test statistic
based on the sample data
Suppose that the sample of 25 cereal
boxes indicates a sample mean 372.5
grams, and sample standard deviation
is 15 grams, so
X   372.5  368
t   1.50
s 15
n 25

01/08/22 lilik for mm 17


5. Make a decision/conclusion
Because the statistic t=+1.50, and
-2.06<+1.50<+2.06, the decision is not to
reject H0, so the mean fill amount is 368 grams
 It means that there is no statistically difference
between 368 grams and 372.5 grams.
 Or, the difference is not significant.
 In other word, the cereal weight of 368 grams and
372.5 grams, mathematically is different, but
statistically that stays the same.math is
deterministic and stat is probabilistic or stochastics.

18 lilik for mm 01/08/22


One - tail test
In many situation, however the
alternative hypothesis focuses on a
particular direction
Note that the alternative
hypothesis contains statement
you are trying to prove

19 lilik for mm 01/08/22


Example
A company that makes processed cheese is
interested in determining whether some
suppliers that provide milk for the processing
operation are adding water to their milk to
increase the amount supplied to the processing
operation. It is known that excess water
reduces the freezing point of the milk. The
freezing point of natural milk is normally
distributed with a mean of -0.5450C. The
standard deviation of the freezing temperature
of natural milk is known to be 0.0080C.

20 lilik for mm 01/08/22


answer
Because the cheese company is only interested in
determining whether the freezing point of the milk is
less than what would be expected from natural milk,
the entire rejection is located in the lower tail of
distribution
Step 1,
 Ho:
  0.5450
H1:
  0.5450

Step 2, LOS = 5%

01/08/22 lilik for mm 21


 Step III

Region of rejection

0.05

-1.71 t

Region of non-rejection

 Reject H0 if t< -1.71; otherwise do not reject H0

22 lilik for mm 01/08/22


Step 4, if a sample of 25 containers of milk is
selected and the sample average freezing point is
determined to be
-0.5500C, so X    0.550    0.545
t   3.125
s 0.008
n 25
Since t= -3.125<-1.71, the observed statistic is in
the region of rejection
Step 5, conclude that the mean freezing point of
the milk provided is below -0.5450C
Empirically: there is an indication that water has
been added to the milk by suppliers.
01/08/22 lilik for mm 23
ex
Berdasarkan organisasi riset kopi, peminumkopi secara
nasional di US mengkonsumsi 3.1 cangkir per hari.
Dari 12 peminum kopi yang diambil secara acak, maka
data nya sbb:
Dengan menggunakan LOS 0.05, apakah ada
perbedaan rata-rata sampel dengan populasi nya?

24 lilik for mm 01/08/22


Rata-rata tingkat bunga pada kredit perumahan sebesar
6%., namun tingkat bunga utk kartu kredit sebesar
14%. Berikut adlh data ttg tkt bunga 10 sampel kartu
kredit.
Apakah rata-rata sampel lebih besar dari 14%?,
gunakan LOS 1%.

25 lilik for mm 01/08/22


Tests about Proportion Population
Proportion: A fraction or percentage
that indicates the part of the
population or sample having a
particular trait of interest.
The sample proportion is denoted by :
number of successes in the sample x
pˆ  
number sampled n

26 lilik for mm 01/08/22


Test Statistic for Testing Population Proportion

ˆ P
p
Z 
P (1  P )
n
P  population proportion
ˆ  sample proportion
p

27 lilik for mm 01/08/22


Example
A study reported in the Wall Street
Journal state that from 899 home
based business, 369 were owned by
women
In the study, the question posed was
“are an equal number of home based
business owned by men and women?

28 lilik for mm 01/08/22


Answer
Step 1
H0: p=0.50 (the proportion of home based
business owned by females is 0.50)
H1:p≠0.50 (the proportion of home based
business owned by females is not 0.50)
Step 2
LOS = 5%

29 lilik for mm 01/08/22


Step 3

Region of rejection H0 Region of rejection


Region of nonrejection

0.95

-1.96 +1.96 Z

30 lilik for mm 01/08/22


Step 4
369
pˆ   0.41046
899
pˆ  P 0.41046  0.50
Z    5.37
P 1  P  0.501  0.50 
n 899
Step 5, conclusion
Because -5.37<-1.96 reject H0. There is
evidence that the proportion of home based
business owned by females is not 0.50

01/08/22 lilik for mm 31


Contoh
Produsen motor memperkirakan
jarak tempuh per liter sepeda motor
“X” 30 km. Hasil tes thd sampel 20
sepeda motor diperoleh data rata-rata
jarak tempuh 28 km dengan standar
deviasi 4 km. Lakukan pengujian
dengan  = 5%
32 lilik for mm 01/08/22
33 lilik for mm 01/08/22
Contoh
Pimpinan perusahaan perusahaan bola
lampu berpendapat daya tahan bola lampu
merek “X” lebih dari 150 jam. Untuk
membuktikan pendapat tersebut diambil
sampel random 25 bola lampu untuk diuji.
Hasilnya rata-rata daya tahan bola lampu
tsb 155 jam dengan simpangan baku 15
jam. Dari data tersebut berilah kesimpulan
ttg pendapat tsb? Gunakan LOS 0,1.

34 lilik for mm 01/08/22


35 lilik for mm 01/08/22
36 lilik for mm 01/08/22
Contoh
Manajer produksi mengatakan
tingkat kerusakan barang kurang dari
10%. Dari 40 sampel random yang
diuji, 3 diantaranya rusak. Benarkah
pendapat tersebut ? Gunakan tingkat
keyakinan 1%

37 lilik for mm 01/08/22


38 lilik for mm 01/08/22
Contoh
“6 dari 10 ABG di Surabaya
merokok”, demikian pendapat
Sosiolog. Hasil penelitian terhadap
100 sampel ABG, 65 diantaranya
merokok. Uji pendapat tersebut
dengan taraf signifikasi 5%

39 lilik for mm 01/08/22

You might also like