File grafting in market research (Q2711717)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: File grafting in market research |
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | File grafting in market research |
scientific article |
Statements
25 April 2001
0 references
singular value decomposition
0 references
supplementary projection
0 references
bootstrap
0 references
procrustean rotation
0 references
File grafting in market research (English)
0 references
0.72960824
0 references
0.7113695
0 references
0.70527184
0 references
0.70298606
0 references
File grafting consists of putting on a single factorial display information from two independent surveys. Classically, these surveys are analysed separately, but very often they contain a common group of variables, for instance, socio-economic variables in market research. Usually, the association pattern of this common group of variables is very stable from one survey to the other. The objective of file grafting is to take advantage of this fact in order to define a common base for the two surveys, to represent the whole information.NEWLINENEWLINENEWLINEIn particular, in market research, this work deals with the association between non-common variables coming from both surveys. This is a difficult and challenging problem since no direct measurement of association is available. Conditions for tackling this problem are studied. A statistical methodology, which is called here as file grafting, is presented to visualize information coming from two independent data sets. For this purpose it is necessary that both data sets share a certain group of variables defining a common space which act as a hinge between the two data sets. Moreover, certain conditions should be fulfilled to control the whole process and interpret the issued results. This methodology is applied for testing the efficiency of a TV advertising campaign from two different panels of households: one measuring the consumption of fast-moving goods and the other measuring the TV audience.
0 references