I am trying to create a workflow that could be used in my data. Let’s say I have 3 columns: Type, Measured value and Theoretical value. I would like to group the first column “Type” and apply one sample t-test. In the “single sample t-test” as a test value I would like to have every theoretical value and for distribution I would like to include measured value. At the end, I need a table where I could have an additional two columns with p-value and t-value as an outcome.
Is there anyone who could help me out with this issue? I have enclosed a provisional workflow
thanks for the provisional workflow, this helps a lot to help
I think this is already very close to what you wanted - basically only the inner loop over the theoretical values was missing (and there where some whitespaces in the group column in the “Table Creator” which I removed with a String Manipulation node):
which you can clean up with a column filter or table manipulator node.
Which KNIME Version are you on? Judging from the Joiner, which shows as deprecated for me, its a rather old version - do you have the possibility to upgrade? I hope the workflow I created works for you then:
Hello Lukas,
Thank you for your response. I am working in the older KNIME version as per requirement of my supervisor Anyway, your workflow helped me a lot. Thank you!
Hello Lukas,
I’ve just realized that I didn’t phrase my question correctly. In the “single sample t-test” as a test value I would like to have every theoretical value that belongs to the same type. At the moment, the workflow calculates with every theoretical values in the table. How can I quickly change that?
Thank you in advance.