Data Quality, Level of Aggregation, and Nonmetric Multidimensional Scaling Solutions

AS THE RELATIVE EMPHASIS IN NONMETRIC MULTIDIMENSIONAL SCALING RESEARCH SHIFTS FROM TECHNIQUES TO APPLICATIONS, INCREASING ATTENTION IS BEING PAID TO THE QUALITY OF INPUT DATA AND THE EXTENT OF AGGREGATION OF INDIVIDUAL DATA FOR ANALYSIS PURPOSES. THE THEORETICAL FRAMEWORK DEVELOPED IN THIS STUDY SH...

Full description

Saved in:
Bibliographic Details
Published inJournal of marketing research Vol. 13; no. 1; pp. 92 - 97
Main Authors Day, George S., Deutscher, Terry, Ryans, Adrian B.
Format Journal Article
LanguageEnglish
Published Chicago American Marketing Association 01.02.1976
SAGE PUBLICATIONS, INC
Subjects
Aggregation
Brands
Breweries
Correlation coefficients
Determinacy
Dimensionality
Effectiveness
Error rates
Judgment
Laboratories
Marketing
Methods
Multidimensional scaling
Research Notes and Communications
Sample mean
Studies
Online AccessGet full text

Cover

More Information
Summary:AS THE RELATIVE EMPHASIS IN NONMETRIC MULTIDIMENSIONAL SCALING RESEARCH SHIFTS FROM TECHNIQUES TO APPLICATIONS, INCREASING ATTENTION IS BEING PAID TO THE QUALITY OF INPUT DATA AND THE EXTENT OF AGGREGATION OF INDIVIDUAL DATA FOR ANALYSIS PURPOSES. THE THEORETICAL FRAMEWORK DEVELOPED IN THIS STUDY SHOULD ASSIST A MARKETING-RESEARCHER WHO HAS NONMETRIC SCALING DATA FROM A LARGE SAMPLE OF RESPONDENTS TO DECIDE WHETHER HE SHOULD SEGMENT HIS SAMPLE, OR WHETHER HE SHOULD SIMPLY WORK WITH THE FULLY AGGREGATED DATA SET. IF HE HAS A VERY LARGE SAMPLE OR HIGHLY RELIABLE DATA BUT HIGH ERROR IN THE AGGREGATE DATA THEN IT WOULD APPEAR THAT THE SAMPLE OF SUBJECTS IS VERY HETEROGENEOUS. IN THIS CASE, SEGMENTATION SHOULD BE TRIED IN AN ATTEMPT TO UNCOVER MORE HOMOGENEOUS SEGMENTS FOR ANALYSIS. THE IMPORTANCE OF COLLECTING RELIABILITY INFORMATION IS DEMONSTRATED WHEN ANY SIMILARITY DATA ARE GATHERED AND ESPECIALLY WHEN AGGREGATION OF THE DATA IS BEING CONSIDERED.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0022-2437
1547-7193
DOI:10.1177/002224377601300114