Q for Cluster Analysis and Latent Class Analysis
Easily perform cluster and latent class analysis
Whether you want to use the familiar cluster analysis techniques or state-of-the-art latent class analysis, it’s easy and efficient to do it in Q.
All the key techniques
Hierarchical clustering, k-means clustering, latent class analysis, and all the associated techniques for predicting class membership in new data sets (e.g., regression, machine learning).
Deal with all data types
Q can be used to form segments with any type of data. Numeric. Categorical. Ranking. MaxDiff. Conjoint. Even text.
Fast to use
Drag and drop variables to create a cluster analysis and latent class analysis.
If you’re an expert you know that missing data can entirely derail cluster and latent class analysis, and simple techniques, such as mean replacement, lead to the results being wrong. Q’s tools automatically deal with missing data problems using the best-practice MAR assumption, so you can focus on strategy and interpretation rather than trying to get the analysis working.
Efficiently profile and compare alternative analyses
Q’s not just a clustering and latent class analysis tool. It is designed to do all your analysis and reporting. This makes it a snap to quickly profile and compare segments, whether you like to do this using crosstabs or some of our various visualizations designed specifically for comparing segments.
Complete (not just factor analysis)
Q is a general-purpose analysis app that does everything from crosstabs to text coding to advanced analysis, driver analysis, and segmentation.
Once you have created your factor analysis, you can use the factors as inputs to all your other work (e.g., crosstabs, regression).
Automate manual tasks and create re-usable analysis and reports.
Find deeper insights
DIY advanced analysis and visualizations.
Manipulate your data with a couple of clicks.
Before Q, market research was a very laborious process. Now I couldn’t do my job without it.
NICOLA SCOTT, NICOLA SCOTT CONSULTING
This content was originally published here.