I have a complex survey that has many question/answer types as well as comments(free form text). I could not find a tutorial how to use the Categorical function in the Consumer Research. It looks like depending on question types how to analyze, but it is not apparent to me.
Good idea Tx,,,, and yes that was the first place I went. I have a survey with 195 columns with questions within questions (attached sample data)). It is chock full of Likert, multiple answers can be checked, as well as demographic, and free form commentary.
I think I just need to map which question type (variables) get what Response role and assign them.
Thanks for reaching out.
Not sure what you want? It takes a good deal of thought to create an effective survey. The creation of the survey (questions, how they are formulated, how many ask the "same" question to get within survey measurement error, how to mitigate bias, etc.) is more important than the analysis itself. You can't fix a poor survey with statistics. That being said, I suggest you read Deming on Survey Errors (The American Statistician, February 2006, Vol. 60, No. 1), Lovitt also has a nice readable paper USING QUALITY TOOLS AND METHODS TO REDUCE BIAS IN JUDGMENT (Quality Engineering, 8(1), 93-116 (1995-96). I don't know the purpose of the survey, but there are plenty of good papers specific to topic.
Regarding the free form text analysis. JMP and JMP Pro have several exploratory to modeling methods for free form text data. My colleague @nickshelton has a nice on demand Mastering JMP webinar which I think can provide you with some great ideas for analyzing the text data in the survey. Here's a link to Nick's webinar:
There are no labels assigned to this post.