There are many forms of data analysis used to report and study research data. Factor analysis is best used to simplify complex data sets with many variables.

What is factor analysis?

Factor analysis is a way of combining the data of many variables into several variables. For this reason, it is also sometimes called "measurement reduction".

You can reduce the "dimensions" of your data to one or more "variables". The most common method is known as component analysis (PCA).

How can factor analysis help?

Let's say you ask a few questions, all of which are related to different but closely related aspects of customer satisfaction:

1. How satisfied are they with your product?
2. Would they recommend your product to a friend or family member?
3. How likely are they to get your product in the future?

But you want only one variable to represent customer satisfaction ratings. One option would be to average the three answers to the questions.

Another option is to create a dependent variable. This can be done by running PCA and saving the first core component (also known as factor).

The advantage of the PCA over the average is that it automatically weighs each of the variables in the calculation.

Let's say you have a list of questions, and you don’t know exactly which answers will move together and which ones differently. For example, barriers to buying potential customers. Possible barriers to buying are listed below:

1. Product Cost.
2. Total implementation costs.
3. Competitor Product Selection.
4. Product benefits do not outweigh the cost.
5. Lack of required features.

Factor analysis may reveal trends in how these issues move together.