Summary advanced management lecture 5
Chapter 4; Exploratory factor analysis
Motivation to use a factor analysis:
Marketing data often have many variables (practical 12 questions)
It is advantageous to reduce these to smaller sets of variables (factors or components)
Seek underlying unobservable (latent) variables that are reflected in the observed variables
(manifest variables)
For instance: a consumer survey with a lot of items (questions) that reflect a smaller number of
underlying concepts such as customer satisfaction, brand preferences or price sensitivity
Latent variables: room quiet dark, you can’t observe anything. There is a object in de middle of the
room. But you have a camera and via the pictures you can observe the object in the middle in the
room. (from different sides)
Basis concept of factor analysis:
Factor analysis: Class of procedures used for data reduction and to summarize the data
(analysis and interpretation of the correlation between the observable variables of a data set)
Factor: a variable or construct that is not directly observable but that needs to be inferred from the
input variables.
Purpose of factor analysis:
1. Identification of underlying constructs (non-observed/latent variables)
2. Reduce the number of variables to a more manageable set
3. Attempt to retain as much of the information as possible and make the remaining variables
meaningful and easy to work with
Assumption: the correlation between the
variables is caused by an underlying factor
The factor is a latent variable which captures
the joint meaning of the items related to it
Chapter 4; Exploratory factor analysis
Motivation to use a factor analysis:
Marketing data often have many variables (practical 12 questions)
It is advantageous to reduce these to smaller sets of variables (factors or components)
Seek underlying unobservable (latent) variables that are reflected in the observed variables
(manifest variables)
For instance: a consumer survey with a lot of items (questions) that reflect a smaller number of
underlying concepts such as customer satisfaction, brand preferences or price sensitivity
Latent variables: room quiet dark, you can’t observe anything. There is a object in de middle of the
room. But you have a camera and via the pictures you can observe the object in the middle in the
room. (from different sides)
Basis concept of factor analysis:
Factor analysis: Class of procedures used for data reduction and to summarize the data
(analysis and interpretation of the correlation between the observable variables of a data set)
Factor: a variable or construct that is not directly observable but that needs to be inferred from the
input variables.
Purpose of factor analysis:
1. Identification of underlying constructs (non-observed/latent variables)
2. Reduce the number of variables to a more manageable set
3. Attempt to retain as much of the information as possible and make the remaining variables
meaningful and easy to work with
Assumption: the correlation between the
variables is caused by an underlying factor
The factor is a latent variable which captures
the joint meaning of the items related to it
