SHARE THE ARTICLE ON
Factor analysis, also known as dimension reductions, is a statistical method of reducing data of larger volume to a smaller data set. As the name suggests, factor analysis basically reduces the dimensions of your data and break it down into fewer variables. This small data set is now more manageable and easy to understand.
Factor analysis finds a repeating pattern in a dataset and observes the common characteristics in the patterns. Hence the “factor” refers to observed variables sharing similar responsive patterns. There are major two types of factor analysis:
Create an actionable feedback collection process.
Factor analysis works by measuring both unique and shared variances across the variables. But to be focused on one, we only consider the variables that share their variances with the other variables and not the unique variables. This is then used to identify various patterns in the variables and group them accordingly.
Factor analysis basically creates a new scale by eliminating the unique variables or the ones that do not share their variances with some other variables. Along with its many applications, factor analysis is commonly used by survey researchers where they want to know whether they can simplify their question responses into shorter ones.
Get market research trends guide, Online Surveys guide, Agile Market Research Guide & 5 Market research Template
Factor analysis helps you in condensing variables and uncovering clusters of responses. Let us look at how it actually works with an example scenario:
You decide to ask some questions like below that cover somewhat similar grounds of customer satisfaction –
As for evaluating the entire performance of the organization, it would be convenient for you to have only one variable that represents the customer experience score. It can be done in two ways:
PCA technique is generally more effective than the average method as it calculated the weightage of the variables along with the calculations.
There might be a case where you want to ask several questions to the customer but don’t exactly know which responses can be grouped together and which will be kept totally different. Example: purchase barriers of the target customers. The reasons for the same could be:
Factor analysis helps you to group these responses like:
We have three labels for the groups and according to them, factor analysis uses heat maps to groups the factors clearly to tell us what affects the responses and what they imply.
Clustering the above information with its three components will tell us the customer trends that are high in Cost and Org but not both.
After all the miracles that factor analysis does, it takes a good observational skill to figure out which questions need factor analysis and which don’t. to make it easier for you, in this section we will be looking at some sample questions where the factor analysis can be a good fit:
These are the “Agree – Disagree” questions that tell the opinions of the customers.
I feel the importance to be loyal to one brand
I value quality more than money
These “Agree – Disagree” questions bring out the behaviour of the customers.
I prefer the costliest option
I am a bargaining customer
These questions measure the attitudes of the customers in an “Agree – Disagree” manner.
I did not like the customer service
I am pleased with the product price
These “Agree – Disagree” questions tell you what the customer usually does.
I always prefer shopping online
I love to visit the zoos
Brand positioning definition: Definition, Types and Benefits SHARE THE ARTICLE ON Share on facebook Share on twitter Share on linkedin Table of Contents When you
DG Solutions expanded its dedicated Healthcare Call Center, DG Connect, using the Voxco survey call center technology. The DG Connect Healthcare Call Center runs outreach and