Name: Student number:
Exam Marketing Research Methods
Friday, April 10 2015
TEACHER
NAME: NUMBER:
FIRST TIME
RESIT for YEAR: BLOCK:
The exam contains XX pages.
The exam consists of 2 parts with several sub-questions.
Please answer each question in the corresponding answer box.
You have 3 hours to complete the exam (14:00-17:00 hrs).
Make sure that you write your name and student number on the front page!
Please answer in English (Dutch is not allowed by the faculty).
Remember that we can only give you points for what you have actually written, not for what
you may have intended to write.
The grades for the exam will be published on Nestor.
Using the sanitary facilities during the exam is not allowed by the faculty.
Number of points per part:
1. AN(C)OVA, Regression, Conjoint Analysis, Moderation/Mediation, Binary data : / 55
2. Factor analysis, Reliability analysis, Cluster analysis: / 45
1
,Name: Student number:
Part 1.
AN(C)OVA, Regression, Conjoint Analysis, Moderation/Mediation, Binary data
(55 points)
1. ANOVA and ANCOVA. IF YOU ARE TAKING A RESIT FOR YEAR 2013-2014 OR BEFORE, YOU
CAN SKIP THIS QUESTION.
A first step one has to take when engaging in an ANCOVA analysis, is the so-called test for
homogeneity of slopes.
a. What is this test about and why do we carry out this test? (2 points)
This test checks whether the effect of the covariate on the dependent variable is independent of the
categorical variable. This should be the case.
If we reject the null hypothesis of homogeneity, we cannot interpret the main effects, and hence
cannot judge the main effect of our categorical variable.
b. Below (next page) you can find the output of such a test. The study that was carried out
investigated whether companies which were presented as more authentic in their CSR
(Corporate Social Responsibility) activities (Authentic = 1 when authentic; = 0 when NOT
authentic) had a better brand identity (higher score on a rating scale). The study accounted
for two covariates, namely Support (how important respondents think CSR activities are) and
Skepticism (i.e. how credible respondents think companies are communicating on their CSR
activities).
What are your conclusions for the homogeneity of slopes test? (2 points)
A the interactions of authentic*support and authentic*skept are NOT significant, so there is
homogeneity of slopes and one can proceed with the ANCOVA analysis without these
interactions
2
, Name: Student number:
Type III Sum of
Source Squares df Mean Square F Sig.
Corrected Model 111,522a 5 22,304 16,726 ,000
Intercept 72,413 1 72,413 54,302 ,000
Authentic ,066 1 ,066 ,050 ,824
Support ,547 1 ,547 ,410 ,523
Skepticism 4,190 1 4,190 3,142 ,078
Authentic * Support 2,734 1 2,734 2,050 ,154
Authentic * Skepticism ,393 1 ,393 ,295 ,588
Error 249,369 187 1,334
Total 3896,000 193
Corrected Total 360,891 192
a. R Squared = ,309 (Adjusted R Squared = ,291)
2. Beta or standardized beta?
The marketing manager of XBrands carried out a regression analysis to find out what the impact is of
1) advertising, 2) price, and 3) sales in previous period on the sales of his premium brand XCluz-If.
The variables have been measured in the following way:
- Sales: volume sales, expressed in millions of units
- Sales in previous period: volume sales, expressed in millions of units
- Advertising: advertising spending, expressed in millions of euros
- Price: unit price, expressed in euros per volume unit
The resulting regression output can be found below.
3