I have the results of a form that has questions and answers for 5 different companies. The company is the header with the questions #s as the rows. What is the easiest set of nodes to use to understand the trends, patterns from these choice responses from these companies?
Hi @aehrenworth
It depends…
Frist read the data into KNIME
Next, do some quality checks . Explore your data, Make some crosstabs.
Are there missing values (and why) Get familiar with your data.
Does the outcome of your exploration makes sense?
Ask yourself: is this data suitable to answer my question(s)?
What does your data looks like and what question do you want answer?
gr. Hans
2 Likes
Thanks for the follow-up. I have done some of these initial steps and data clean-up -. Just looking for any nodes that might help to identify similarities/differences between how certain companies answer the 30-40 questions. Categories also organize the questions, so that is an element that could be useful as well.
appreciate the help
Its difficult to offer help without knowing more about the data and specifically what you want to do. For example - Are the answers nominal or numeric? Is there any reason to weight some questions more than others? Are you trying to compare the companies? You mention trends - is there a time element to the answers? Some anonymized data would be helpful.
1 Like