Round 2 Lecture 5 Flashcards Preview

Marketing Research > Round 2 Lecture 5 > Flashcards

Flashcards in Round 2 Lecture 5 Deck (14)
Loading flashcards...

Why Segment the Market?

-A basic belief in Marketing is to deliver the right products, to the right people, in the right way
-If that is done successfully, one will be able to attract a greater price premium and one will be able to retain customers better.


Major Segmentation Variables

 Geographic
 Purchasing Approaches
 Personal Characteristics (psychographics)



Measure consumer attitudes (needs, preferences)


factor analysis

 Find latent psychological drivers


Use cluster analysis to identify prototypical customer profiles (segments)

 In some cases, product offering can be optimized for each cluster
 Communications can be designed to appeal to particular clusters


factor analyisis

-Develop factors, also called “super-variables”, “meta- attributes” or “general constructs” that describe consumer attitudes, beliefs. Etc.
- Reduce redundancies in survey items
- Create uncorrelated “super variables” for regression,
clustering, etc.


The objective of the factor analysis is to

represent each of the original variables (X) as a linear combination of a smaller set of factors (F)


Goal of factor analysis

To reduce the number of variables and identify underlying constructs (i.e., factors) that can be managed and are actionable


Input of factor analysis

The responses to your survey questions for each respondent


 Output of factor analysis

(1) The factor loadings and variance-explained percentages,
(2) a set of independent factors scores (the “new Xs”)


Steps for Factor Analysis

1. Decide on the number of factors
2. Derive the factor solution
3. Interpret each of the factors
4. Evaluate the quality of the fit
5. Save the factor scores for use in subsequent analyses


Rotation is

a transformation of the initial solution into a new solution which is easier to interpret
• Orthogonal rotation (varimax, quartimax, minimax)
• Oblique rotation


The % of variance explained by your K factor solution

gives you an indication of the quality of the overall fit.


K-means clustering is

-the most commonly used clustering technique for segmentation.
- It is an iterative technique that seeks to allocate each observation to the cluster that is located closest to it
- The number of clusters is chosen by you, and you decide upon the correct number of clusters using either business criterion, or by statistical criterion.