Scanner Dataset

Essay by SardaoUniversity, Master'sA+, November 2004

download word file, 25 pages 5.0

Table of Contents

Table of Contents 1

Executive summary 2

Purpose and scope of document 2

Methods 2

Results and Conclusions 2

Recommendations 3

Evaluate the design and data collection plan for the scanner study. 4

Descriptive information of the sample population 4

Description of the brand purchasers 8

Consumer of brand A: 10

Consumer of brand B: 11

Consumer of brand C: 11

Consumer of brand D: 11

Appropriate statistical tests 12

Identification of the best model that should be used for predictive purposes 12

Linear Regression 12

Discriminant Analysis 13

Logistic Regression 17

Comments on the variables that are the best predictors of buying behaviour 19

Linear Regression Model: 19

Logistic Regression Model: 19

What are the marketing strategy implications of your analysis in item 2? 20

Use the model found in item 2.2 and determines which brand is expected to be purchased 21

Linear Regression Model 21

Logistic Regression Model: 21

Analysis of the results: 22

Relations between variables in X2 22

Results and Conclusions 25

Recommendations 25

Executive summary

Purpose and scope of document

The purpose of this research was to find a model that would be able to predict buying behaviour of specific set of consumers towards four different brands.

Given the Dataset available, we tried to define the costumers for each brand and try to segment them so that a better prediction would result and to be able to provide useful and consistent managerial recommendations.


The research was first conducted through a descriptive study, with the data set collected trough a scanner. Moreover the scanner panel data has a household identifier containing information about demographics, tv and newspaper habits and other variables. The scanners used were placed in Points of Sale (grocery stores).

We used SPSS in its full capacities with the...