The factorial analysis has undergone numerous developments and calculation methods. Although it was used for the first time by specialists in psychometrics, its field of application has been extended to many other fields, such as geology, medicine and finance. factor analysis in research
Factor analysis is a statistical technique that is used today mainly to analyze surveys . It allows, when we have a population of individuals of whom we have a lot of information regarding opinions, practices and status (sex, age, etc.), to see the similarities and oppositions between the characteristics of the individuals.
For example, when applying a satisfaction survey you are going to collect a large number of variables at the same time. We want to know if the questions in the questionnaire are grouped in any characteristic way. Through this analysis we can find groups of variables in the responses of the respondents that have a common meaning and thus reduce the number of dimensions necessary to explain the responses received.
Let’s learn more about its characteristics and the types that exist.
What is factor analysis?
Factor analysis is one that reveals the relationships between variables and, in general, understand the data that is modeled.
This analysis is part of the generalized linear model (GLM) and this method also assumes several assumptions: there is a linear relationship, there is no multicollinearity, it includes the relevant variables in the analysis, and there is a correlation between the variables and the factors.
Factor analysis is a useful tool to investigate the relationships between variables of complex concepts such as socioeconomic status or psychological scales.
It allows to investigate concepts that are not easy to measure directly, collapsing a large number of variables into a few interpretable underlying factors.
Methods for a factor analysis
There are different types of methods used to extract the factor from the data set:
- Principal component analysis: It is the method most used by researchers. You start by extracting the maximum variance and placing it in the first factor. Then you remove the variance explained by the first factors, and then begin to extract the maximum variance for the second factor. This process goes down to the last factor.
- Common analysis : The second most preferred method by researchers, extracts the common variance and factor it. This method does not include the unique variance of all variables.
- Image factoring: This method is based on the correlation matrix. The regression method is used to predict the factor in the image factorization.
- Maximum similarity method : This method also works on the correlation metric but uses the method of maximum similarity to factor.
Factor analysis models
Let’s know some of the characteristics of the types of factor analysis:
- Exploratory factor analysis: It assumes that any indicator or variable can be associated with any factor. It is the factor analysis most used by researchers and is not based on any previous theory. factor analysis in research
- Confirmatory factor analysis : It is used to determine the factor and factor load of the measured variables, and to confirm what is expected in the basic or pre-established theory. The CFA assumes that each factor is associated with a specific subset of measured variables. It usually uses two approaches:
Factor analysis applies, for example, to make the buying decision process of a customer. You can analyze different types of variables, from emotional to economic and social for a better interpretation of the results.