1
Lecture 9
Hypothesis Testing Related to
Differences (and some tips about
assignment 2)
Department of Marketing
Dr. Junzhao Ma
MKF2121
Marketing Research Methods
2
Today’s agenda
3
•
Hypothesis testing – tests of differences
–
Different types of t tests
–
One-way ANOVA
•
How to write about hypothesis testing
•
A few words on assignment 2
Quick review of week 8 material
•
Hypothesis testing
•
Cross tabulation/chi-
square test
4
Key Concepts
•
Sample vs. population
characteristics
•
Null vs. alternative
hypothesis
•
Test statistic
•
P Value
•
Cross
‐
tabulation/chi
‐
square test
2
The procedure of hypothesis testing is the same for different
RQs. What is different is 1. the test you use and 2. how to
interpret the results if you reject the null hypothesis
5
1. Problem Definition (RQ)
2. Clearly State the
Null and Alternative Hypotheses
3. Choose the
relevant statistical test
4. Calculate p-value
5. Is p-value<0.05?NoDo not reject null
1. One Sample
2. Two Independent
Samples
3. Paired
Samples
4. More Than
Two Samples
The different types of tests of differences
t tests
ANOVA
7
3
All the tests of differences are about comparing
average
values
8
One sample t test
•
Compare the average value of one variable to a
constant
•
Example RQs
– Is the mean service rating above 5?
– Is average customer spending equal to $500?
– Is the average miles per gallon (fuel efficiency
measure) equal to 30?
17-9
Paired samples t test
•
Compare the average value of two variables
(from one group of respondents)
• Example
– Is the mean consumer rating for McDonalds different
from that of Hungry Jacks?
– Is consumers’ mean willingness to pay for iphone 6
different from that of Samsung S6?
17-10
4
Two independent samples t test
•
Compare the average value of one variable
from two different groups
•
Example RQs
– Does the mean rating for the new iphone 7 differ
between users and non-users?
– Does the mean purchase intention differ between
men and women?
17-11
Which t test to use?
1 Group
2 Groups
1 metric Variable
One sample t test
H
0
: pop mean = value
Two Independent
samples t test
H
0
: pop mean for 1
st
pop =
pop mean for 2
nd
pop
2 metric Variables
Paired Sample t test
H
0
: pop mean var 1 = pop
mean var 2
17-12
Why are they all called t tests?
t statistic
P value* = the
shaded area
Because their test statistic follows a t distribution
17-13
5
Hypothesis Testing Using the t Statistic – same as
before
17-14
1. Problem Definition (RQ)
2. Clearly State the
Null and Alternative Hypotheses
3. Choose the relevant statistical test
(one of the t tests)
4. Calculate p-value
