Exam Marketing Research Methods
Wednesday, November 9 2016
TEACHER
The exam contains 28 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 (18.30-21.30 hrs).
Make sure that you write your name and student number the first page!
Please answer in English (Dutch is not allowed by the faculty).
The grades for the exam will be published on Nestor.
We wish you lots of success!!!
Name
Student number
Year / semester you followed MRM
Maximum number of credits per question:
Part 1: AN(C)OVA & different types of Regression (55 credits)
Part 2: Factor, Reliability & Cluster (45 credits)
1
,Part 1. AN(C)OVA, regression, conjoint, moderation and mediation, binary regression (55 credits)
In order to analyze the effect of temperature on sales of Alps full-milk chocolate, a marketing analyst
generates a simple dummy variable with the following levels:
Variable name levels
tmpwarm =1 if temperature > 17 , =0 otherwise
After that, she checks for the difference in sales for the two different temperature conditions, using an
ANOVA, resulting in the following tables.
Descriptives
sales
95% Confidence Interval
for Mean
Std. Lower Upper
N Mean Deviation Std. Error Bound Bound Minimum Maximum
,00 55 130,738 105,898 14,279 102,110 159,367 60,359 530,874
1,00 13 113,534 75,582 20,962 67,860 159,208 68,730 359,070
Total 68 127,449 100,539 12,192 103,114 151,785 60,359 530,874
ANOVA
sales
Sum of Squares df Mean Square F Sig.
Between Groups 3112,207 1 3112,207 ,305 ,583
Within Groups 674133,306 66 10214,141
Total 677245,513 67
1) Does the sales significantly differ for different conditions? Motivate your answer. (3 credits )
We only have two levels, a t-test would also do the job (0.5 points)
The F-stat is small, and when comparing it to the critical value of under 5% it is not showing a significant
difference in average sales for the two different temperature states.
This can also be seen by looking at Sig. which is above 5%.
This can also be seen by the small difference between Within Groups variance and Total variance.
2
, The managers want also to control for the price to see the effect of temperature. The following ANCOVA
tables pop up.
Between-Subjects Factors
N
tmpwarm ,00 55
1,00 13
Tests of Between-Subjects Effects
Dependent Variable: sales
Type III Sum of
Source Squares df Mean Square F Sig.
a
Corrected Model 608666,301 3 202888,767 189,341 ,000
Intercept 380683,763 1 380683,763 355,265 ,000
tmpwarm * price1 9394,091 1 9394,091 8,767 ,004
tmpwarm 9781,916 1 9781,916 9,129 ,004
price1 298929,125 1 298929,125 278,969 ,000
Error 68579,212 64 1071,550
Total 1781798,706 68
Corrected Total 677245,513 67
a. R Squared = ,899 (Adjusted R Squared = ,894)
2) Given the analysis above, can we isolate the effect of temperature on sales independently of price?
Motivate your answer. (5 credits )
Since price is continuous variable we are performing now an ANCOVA (0.5)
By looking a the interaction term of price and the factorial variable, we can see that it is significant under 5
% level (1.5)
This means that the assumption of homogeneity of slopes, which says that the effect of price on sales is
the same on sales throughout the whole sample, independent of the temperature does not hold! (1.5)
In this case we cannot isolate the effect of price and temperature, and thus we cannot say anything about
their effects individually. We can only say that overall there is a significant effect of price and temp on
sales (1.5)
3
Wednesday, November 9 2016
TEACHER
The exam contains 28 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 (18.30-21.30 hrs).
Make sure that you write your name and student number the first page!
Please answer in English (Dutch is not allowed by the faculty).
The grades for the exam will be published on Nestor.
We wish you lots of success!!!
Name
Student number
Year / semester you followed MRM
Maximum number of credits per question:
Part 1: AN(C)OVA & different types of Regression (55 credits)
Part 2: Factor, Reliability & Cluster (45 credits)
1
,Part 1. AN(C)OVA, regression, conjoint, moderation and mediation, binary regression (55 credits)
In order to analyze the effect of temperature on sales of Alps full-milk chocolate, a marketing analyst
generates a simple dummy variable with the following levels:
Variable name levels
tmpwarm =1 if temperature > 17 , =0 otherwise
After that, she checks for the difference in sales for the two different temperature conditions, using an
ANOVA, resulting in the following tables.
Descriptives
sales
95% Confidence Interval
for Mean
Std. Lower Upper
N Mean Deviation Std. Error Bound Bound Minimum Maximum
,00 55 130,738 105,898 14,279 102,110 159,367 60,359 530,874
1,00 13 113,534 75,582 20,962 67,860 159,208 68,730 359,070
Total 68 127,449 100,539 12,192 103,114 151,785 60,359 530,874
ANOVA
sales
Sum of Squares df Mean Square F Sig.
Between Groups 3112,207 1 3112,207 ,305 ,583
Within Groups 674133,306 66 10214,141
Total 677245,513 67
1) Does the sales significantly differ for different conditions? Motivate your answer. (3 credits )
We only have two levels, a t-test would also do the job (0.5 points)
The F-stat is small, and when comparing it to the critical value of under 5% it is not showing a significant
difference in average sales for the two different temperature states.
This can also be seen by looking at Sig. which is above 5%.
This can also be seen by the small difference between Within Groups variance and Total variance.
2
, The managers want also to control for the price to see the effect of temperature. The following ANCOVA
tables pop up.
Between-Subjects Factors
N
tmpwarm ,00 55
1,00 13
Tests of Between-Subjects Effects
Dependent Variable: sales
Type III Sum of
Source Squares df Mean Square F Sig.
a
Corrected Model 608666,301 3 202888,767 189,341 ,000
Intercept 380683,763 1 380683,763 355,265 ,000
tmpwarm * price1 9394,091 1 9394,091 8,767 ,004
tmpwarm 9781,916 1 9781,916 9,129 ,004
price1 298929,125 1 298929,125 278,969 ,000
Error 68579,212 64 1071,550
Total 1781798,706 68
Corrected Total 677245,513 67
a. R Squared = ,899 (Adjusted R Squared = ,894)
2) Given the analysis above, can we isolate the effect of temperature on sales independently of price?
Motivate your answer. (5 credits )
Since price is continuous variable we are performing now an ANCOVA (0.5)
By looking a the interaction term of price and the factorial variable, we can see that it is significant under 5
% level (1.5)
This means that the assumption of homogeneity of slopes, which says that the effect of price on sales is
the same on sales throughout the whole sample, independent of the temperature does not hold! (1.5)
In this case we cannot isolate the effect of price and temperature, and thus we cannot say anything about
their effects individually. We can only say that overall there is a significant effect of price and temp on
sales (1.5)
3
