Marketing Research
Session: January 18, 2010
Instructors (in alphabetical order):
Prof. Dr. Els Gijsbrechts
Drs Mark Vroegrijk
Session Overview
Course Objectives/Contents
Course Organization/Practical Issues
Introduction (HBBAT: Chapter 1)
1. Course objectives/contents
Course Objectives
After this course you should be able to judiciously use (Multivariate) statistical methods to support strategic and tactical marketing decisions
Support Strategic and Tactical Marketing Decisions?
Segmentation and Targeting
Positioning
Guide Marketing Mix Decisions:
Pricing
Distribution
Communication
Product/Assortment
Some examples of Marketing Research Problems :
Villa Pardoes, Esprit, Tio-Pepe Sherry: Image/Competitive Positioning?
Dela, Océ, T-Mobile: Optimal ‘product package’?
Albert Heijn/Vodafone: Effectiveness of local promotion ads / in-store electronic displays?
Neckermann mail order company: Set up efficient direct mailing campaigns?
ixX Pharma: Improve Shelf Display?
Fortis Bank: Segmentation and Targeting the ‘private’ market?
Example: Positioning
Professional
Experts
Cheap
Fast
Postbank
ABN
ING
RABO
Fortis
Course objectives: Develop …
Knowledge:
Theoretical : know objectives, principles and assumptions of selected multivariate methods
Marketing: Be able to identify proper method to solve marketing problem on hand
Lecture sessions
Skills: be able to apply multivariate methods/solve marketing problems using SPSS
-> SPSS instruction session and assignment
Course position in curriculum
Complementary course : ‘Methoden van Bedrijfseconomisch onderzoek’ (‘Research Methods in Business’): outlines basic principles of research process (including data collection and (univariate) analyses) for solution of business problems
Marketing Research: Focus on (i) Marketing Problems, (ii) Quantitative Mainstream Multivariate Analyses
Optional follow up courses: Marketing Models, Advanced Marketing Research, Survey Methodology, Experimental research
2. Course Organization /Practical Issues
Overview of Sessions/ Time Table see Reader on BB
Lectures:
Intro
Per Topic/Method:
Principles
Case with SPSS instructions
Guest Lectures
Assignment (incl instruction session)
Course Materials
Lecture Sheets/Notes: see BB
Book: Hair, Black, Babin, Anderson, Tatham (2006)=HBBAT, Multivariate Data Analysis (Specified Chapters )
Selected Reading: see BB
Practical Issues
Assignment
In groups of 4 to 5 students
Due date: see later
‘Office hours’ for Q&A and feedback
Assignment questions/feedback:
M. Vroegrijk
Grading
Written exam (70 %)
Assignment (30 %)
15 –20 pages +output
Students should obtain at least 5(on 10) for each part in order to pass the course
Assignment grade transfer: WRITTEN request, see announcement BB
Closure
Exam: May 17, 2010
Re-exam: July 9, 2010
Introduction
HBBAT: Chapter 1
Defining Multivariate Analysis
HBBAT:
‘Broadly speaking, it refers to all statistical methods that simultaneously analyze multiple measurements on each individual of object under investigation’
Why Bother?
Almost every real life marketing problem requires statistical analysis of several variables: your need them in your toolkit!
Crucial for Master Thesis:
‘Translate’ marketing problem
Collect data
Analyse using SPSS
The Power of Multivariate Analysis
Descriptive power
Explanatory power
Predictive power
Normative power/decision support
Strenghts of selected multivariate methods
S
P
S
S
Conjoint
N
S
P
S
Multidim. Scaling/CA
N
N
S
P
Cluster An.
N
N
S
P
Factor An.
N
P
S
S
(M)ANOVA
N
P
S
S
Discriminant en logit
N
P
S
S
Lin. Regres.
Guide (DS)
Predict
Explain
Describe
Method
P=primary, S=secondary, N=not a strength
The Variate
Variate value =
w1X1 + w2X2 + w3X3 + … + wnXn
Data Scales
Nonmetric Scales:
Nominal
Ordinal
Metric Scales:
Interval
Ratio
Nominal Scale
Characteristics:
unique definition/identification
classification
unaffected by one to one transformations
Phenomena:
. brand name, gender, store type
Appropriate Methods of Analysis/Statistics:
.: %, mode, Chi square tests
Ordinal Scale
Characteristics:
indicate 'order', sequence
insensitive to monotonic transformations
Phenomena:
. preference ranking, level of education
Appropriate methods of analysis/statistics:
percentiles, median, rank correlation
+ all previous statistics
Interval Scale
Characteristics:
arbitrary origin
insensitive to linear positive transformations
Phenomena:
. attribute scores, price index
Appropriate Methods of analysis:
arithmetic average, range, standard deviation, product-moment correlation, + previous methods
Ratio Scale
Characteristics:
unique origin
insensitive to linear transformations through origin
Phenomena:
. age, cost, number of customers
Appropriate methods of analysis:
geometric average, coefficient of variation, + all previous methods
Errors: Reliability and Validity
Reliability: Is the measure ,consistent’, correctly geregistered, ..?
Validity: Does the measure capture the concept it is supposed to measure?
Example: Income
Hypothesis testing
Power
Power depends on:
α
Effect size
Sample Size n
Implications:
Anticipate consequences of α, effect and n
Assess/incorporate power when interpreting results
Using Multivariate methods: stages
1. Define Objectives/Purpose
2. Develop the ‘Analysis Plan’
3. Evaluate assumptions/ conduct preliminary data analysis
4. Model estimation and Fit
5. Interpretation of outcomes
6. Validation of results
Using Multivariate methods: some guidelines
statistical significance and practical relevance
‘power’
inspect data
strive for parsimony
examine errors
validate outcomes
Multivariate Methods:
Dependence or Interdependence techniques
Dependence Techniques
One or more variables can be identified as dependent variables and the remaining as independent variables
Choice of dependence technique depends on the number of dependent variables involved in analysis
Interdependence Techniques
Whole set of interdependent relationships is examined
Further classified as having focus on variable or objects
Some examples of Marketing Research Problems :
Villa Pardoes, Esprit, Tio-Pepe sherry: Image/Competitive Positioning?
Dela, Océ, T-Mobile: Optimal ‘product package’?
Albert Heijn/Vodafone: Effectiveness of local promotion ads / in-store electronic displays?
Neckermann mail order company: Set up efficient direct mailing campaigns?
ixX Pharma: Improve Shelf Display?
Fortis Bank: Segmentation and Targeting the ‘private’ market?
What type of relationship is being examined?
How many variables are being predicted?
What is the measurement scale of the dependent variable?
Canonical correlation analysis (chapter 8)
What is the measurement scale of the dependent variable?
What is the measurement scale of the predictor variable?
Multivariate analysis of variance (chapter 6)
Metric
Nonmetric
Metric
Nonmetric
Structural equation modeling (chapter 11)
Canonical correlation analysis with dummy variables (chapter 8)
Metric
Nonmetric
Multiple regression (chapter 4)
Conjoint analysis (chapter 7)
Dependence
Multiple relationships of dependent and independent variables
Several dependent variables in single relationship
One dependent in a single relationship
Multiple discriminant analysis
(chapter 5)
Linear probability models
(chapter 4)
How many variables are being predicted?
How are the attributes measured?
Factor analysis (chapter 3)
Correspondence analysis
(chapter 10)
Metric
Nonmetric
Cluster analysis (chapter 9)
Multidimensional scaling
(chapter 10)
Interdependence
Variable
Cases/
Respondent
Object
Nonmetric
