Latent Class Analysis for User Segmentation


New Member
Hi! I am a UX researcher, and I am trying to use both qualitative and quantitative methods to identify personas, types of users. To that end, I am going to send out a survey to a large sample of our users, identify segments using some statistical clustering technique, determine key variable that differentiate the segments, and predict segment membership using a typing tool. All that sounds very logical to me but also completely latent: I have no former knowledge of statistics.

Some questions:
- I want to make sure that the data I collect (questions I ask on the survey) is going to be conducive to do the LCA. Do I need the answers to my questions in a particular format (numerical?) or does it not matter at all?
- All tools I've seen that can perform the LCA are for Windows and/or costly. Is there a way I can perform the analysis for free using a Mac?
- Is there a less complex tool or method I can use but still add some scientific rigor into the analysis?
- What is a "typing tool"?

Thank you! Any help would be welcomed.