DSC 2320
Statistical Analysis for Management
Chapter 04
Hypothesis Testing
Department of Decision Sciences 1
Session - 05
Hypothesis Testing
(One Sample Tests)
Learning Outcomes
At the end of the chapter, the student should be able to;
Perform the One-Sample t-test using SPSS and interpret
the results
Perform the Binomial test using SPSS and interpret the
results
Perform the Chi-square goodness of fit test using SPSS
and interpret the results
2
Parametric & Nonparametric Tests
Parametric tests Nonparametric Tests
One sample Binomial Test
Tests One Sample t-test Chi-square goodness of fit test
Two Sample 1. Independent Mann Whitney U test
Tests samples t-test
2. Paired sample Wilcoxon signed rank test
t-test for paired samples
Kruskel Wallies H test
1. One-way
Friedman test (ordinal data)
Three or More ANOVA Cochran Q test (Binary data)
samples 2. ANOVA with
Repeated Chi-square test of
Measures independence
3
Testing Hypothesis for Population Mean
The One-Sample t-test procedure
The One-Sample t-test procedure tests the mean
value of a single variable.
H0: µ = 10 Vs H1: µ ≠ 10
H0: µ ≥ 10 Vs H1: µ < 10
H0: µ ≤ 10 Vs H1: µ > 10
4
One sample t-test
Conditions
Data should be observed from a normally
distributed variable.
(Normality of the interested variable should be
assessed before performing the test)
Data should be either interval or ratio
5
The One-sample t-test…
For Two-tail tests
P value = Sig
For One-tail tests
For right-tail tests (i.e. H1: > 0)
If t > 0 then, P value = Sig / 2
If t < 0 then, P value = 1 – Sig / 2
For left-tail tests (i.e. H1: < 0)
If t < 0 then, P value = Sig / 2
If t > 0 then, P value = 1 – Sig / 2 6
One sample t-test
SPSS Command
7
One sample t-test
Case 02 @ page 25
Example 5.1
Test whether the mean value of the area used to
cultivate paddy is significantly different from 3
acres. Write the hypotheses for the test and
perform the test at 5% level of significance.
Data file: [Link]
8
One sample t-test
Case 02 @ page 25
Example 5.2
Test whether the mean value of amount of money
spent on fertilizer is significantly different from Rs.
10,000. Write the hypotheses for the test and
perform the test at 5% level of significance.
Data file: [Link]
9
One sample t-test
Case 02 @ page 25
Example 5.3
Test whether the mean value of amount of money
spent on fertilizer by farmers who cultivate the
improved variety is significantly less than Rs.
9,500. Write the hypotheses for the test and
perform the test at 5% level of significance.
Select the variable by using if case.
10
Data file: [Link]
One sample t-test
Case 07 @ page 26
Example 5.4
Test whether the mean value of absenteeism rate
in June, 2013 is significantly greater than 10%.
Write the hypotheses for the test and perform the
test at 5% level of significance.
Data file: [Link]
11
The Binomial test
This procedure compares an observed proportion
of cases to the proportion expected under a
Binomial distribution with a specified probability
parameter.
When performing a One sample t-test
If the normality assumption is violated(N= SMALL)
If the sample size is not enough
Population median can be tested using the Binomial
test .parameter is “p”
12
The Binomial test…
The Observed proportion
The observed proportion is defined either by the
number of cases having the first value of a
dichotomous (a variable that has two possible
values) variable
or
The observed proportion is defined by the
number of cases at or below a given cut point
on a scale (quantitative) variable.
13
Binomial test…
The binomial test allows us to test whether the
proportion of cases having the first value of a two-
level categorical variable
H0: P = po Vs H1: P po
H0: P ≥ po Vs H1: P < po
H0: P ≤ po Vs H1: P > po
SPSS Command
Analyze Nonparametric Binomial test 14
Binomial Test…
Case 09 @ page 26
Dr. Dharmapriya, the branch manager of
RANMASU bank conducted a study to identify the
factors affecting to savings of its depositors. A
random sample of 500 normal saving depositors
was selected.
Data file: [Link]
15
The Binomial test…
Example 5.5
Test whether the proportion of male depositors is
significantly different from 0.5. Write the
Hypotheses for the test. Perform the test at 5% level
of significance.
Data file: [Link]
16
The Binomial test…
Example 5.6
Test whether the proportion of depositors who have
obtained a bank loan is significantly less than 0.4.
Write the Hypotheses for the test. Perform the test
at 5% level of significance.
Data file: [Link]
17
The Binomial test…
Hypotheses for testing the median
H0: median = mo Vs H1: median mo
Set the ‘Cut Point’ to the hypothesized median
value (m0)
18
The Binomial test…
SPSS Command
19
The Binomial test…
Example 5.7
Test whether the average value of other income of
farmers is significantly different from Rs. 5,000.
Write the hypotheses for the test and perform the
test at 5% level of significance.
Data file: [Link]
Avg is not = mean
20
The Binomial test…
Example 5.8
Test whether the average value of yield in Yala
season of farmers is significantly different from185
Kg. Write the hypotheses for the test and perform
the test at 1% level of significance.
Data file: [Link]
21
Testing hypothesis about one categorical
variable
Chi-square goodness of fit test
The test is applied when you have
one categorical variable
Research Question to be answered
For a categorical variable with k categories, are
the population falling into each of the k categories
as specified?
22
One sample Chi-square test
SPSS Command
23
One sample Chi-square test…
Example 5.9
Test whether the depositors who deal with the bank
is independent from the level of education. Write
the Hypotheses for the test. Perform the test at 5%
level of significance.
Data file: [Link]
24
One sample Chi-square test…
Example 5.10
Test whether the customers of YMB shopping
network is independent from the method of
payment. Write the Hypotheses for the test. Perform
the test at 5% level of significance.
Data file: [Link]
25
One sample Chi-square test…
Example 5.11 Data file: [Link]
According to a survey conducted in the last year, the
probability distribution of reason for shopping of
customers in KLC shopping network is as follows.
Reason Convenience Prices Promotional Other
Probability 0.2 0.6 0.1 0.1
It is required to test whether there is sufficient
evidence to conclude that the distribution of reason
for shopping of customers of YMB shopping network
is different from the above distribution. Write the
Hypotheses for the test. Perform the test at 5% level
26
of significance.
