1
Lecture 8
•
Logic of Hypothesis Testing
•
Cross-Tabulation
Department of Marketing
Dr. Junzhao Ma
MKF2121
Marketing Research Methods
1
Today’s agenda
2
•
More on hypothesis testing – the intuition behind
statistical tests and the concept of a test statistic
•
Learn about one type of tests of association: cross-
tabulation (Chi-square test)
Last week we discussed the steps of hypothesis testing
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?
No
Do not reject null
Reject null hypothesis
Yes
3

2
4
Revisit - why do we have to use statistical tests to “answer” RQ?
•
Sample characteristics
≠
population characteristics
•
Population characteristics is unknown. Sample
characteristics is a
random variable
•
Think like a detective: what does the evidence (sample
characteristics) tell me what really happened (population
characteristics)?
•
Hypothesis testing
: the
formal
process to use the
statistical properties
of the data to evaluate beliefs
about population characteristics (hypothesis)
Population?
Sample
Statistical testing - three levels of understanding
•
Mechanical – know which button to push in SPSS, what to
look for in the output
•
Mathematical (formula) – nice to have (in the textbook),
but not required
•
Conceptual – the intuition behind the technique used
5
Example to illustrate the rationale behind the
statistical testing (in particular, test statistic)
6
Buyer ID
Miles per gallon
1
30.9
2
24.5
3
31.2
4
28.7
5
35.1
6
29
7
28.8
8
23.1
9
31
10
30.2
11
28.4
12
29.3
13
24.2
14
27
15
26.7
16
31
17
23.5
18
29.4
19
26.3
20
27.5
21
28.2
22
28.4
23
29.1
24
21.9
25
30.9
•
Toyota dealership in the city of
New Haven claims that the
average miles per gallon (MPG)
is 30 for the new Toyota Corolla
•
A local consumer group
collected data from the recent
buyers to verify this claim
•
RQ – Is the new Toyota
Corolla’s average MPG equal to
30?